virulence gene and crispr multilocus sequence typing
TRANSCRIPT
The Pennsylvania State University
The Graduate School
Department of Food Science
VIRULENCE GENE AND CRISPR MULTILOCUS SEQUENCE TYPING
SCHEME FOR SUBTYPING THE MAJOR SEROVARS
OF SALMONELLA ENTERICA SUBSPECIES ENTERICA
A Thesis in
Food Science
by
Fenyun Liu
2010 Fenyun Liu
Submitted in Partial Fulfillment
of the Requirements
for the Degree of
Master of Science
December 2010
ii
The thesis of Fenyun Liu was reviewed and approved* by the following:
Stephen J. Knabel
Professor of Food Science
Thesis Co-Advisor
Edward G. Dudley
Assistant Professor of Food Science
Thesis Co-Advisor
Bhushan M. Jayarao
Professor of Veterinary Science
Rodolphe Barrangou
Adjunct Professor of Food Science
John D. Floros
Professor of Food Science
Head of the Department of Food Science
*Signatures are on file in the Graduate School
iii
ABSTRACT
Salmonella enterica subsp. enterica is the leading cause of bacterial foodborne disease in the
United States. Molecular subtyping methods are powerful tools for tracking the farm-to-fork
spread of foodborne pathogens during outbreaks. In order to develop a novel multilocus
sequence typing (MLST) scheme for subtyping the most prevalent serovars of Salmonella, the
virulence genes fimH and sseL and Clustered Regularly Interspaced Short Palindromic Repeat
(CRISPR) regions were sequenced from 171 clinical isolates from serovars Typhimurium,
Enteritidis, Newport, Heidelberg, Javiana, I 4, [5], 12; i: -, Montevideo, Muenchen and Saintpaul.
Another 63 environmental isolates and 70 poultry isolates of S. Enteritidis from poultry industries
in PA were also analyzed. The MLST scheme using only virulence genes was insufficient to
separate all unrelated outbreak clones. However, the addition of CRISPR sequences dramatically
improved discriminatory power of this MLST method. Moreover, the present MLST scheme
provided better discrimination of S. Enteritidis strains than PFGE. Cluster analyses revealed the
current MLST scheme is highly congruent with serotyping and epidemiological data. For the
analyses with S. Enteritidis isolates, the current MLST scheme identified three persistent and
predominant sequence types circulating among humans in the U.S. and poultry and hen house
environments in PA. It also identified an environment-specific sequence type. Moreover, cluster
analysis based on fimH and sseL identified three epidemic clones and one outbreak clone of S.
Enteritidis. In conclusion, the novel MLST scheme described in the present study accurately
differentiated outbreak clones of the major serovars of Salmonella, and therefore may be an
excellent tool for subtyping this important foodborne pathogen during outbreak investigations.
Furthermore, the MLST scheme may provide information about the ecological origin of S.
Enteritidis isolates, potentially identifying strains that differ in virulence capacity.
iv
TABLE OF CONTENTS
LIST OF FIGURES……………………………………………………………………………vi
LIST OF TABLES…………………………………………………………………………….vii
LIST OF ABBREVIATIONS AND DEFINITIONS…………………………………………viii
ACKNOWLEDGEMENTS…………………………………………………………………….x
Chapter 1 Statement of the problem ........................................................................................ 1
Chapter 2 Literature review ..................................................................................................... 3
2.1 Salmonellosis ............................................................................................................. 3 2.1.1 Salmonella ....................................................................................................... 4 2.1.2 Salmonella taxonomy and serotyping ............................................................. 4 2.1.3 Evolution of pathogenicity .............................................................................. 5 2.1.4 Salmonella reservoirs ...................................................................................... 6 2.1.5 Salmonella association with foods .................................................................. 8 2.1.6 Most common Salmonella serovars associated with human illnesses ............. 9
2.2 Subtyping of Salmonella ............................................................................................ 15 2.2.1 Important definitions and performance criteria of subtyping methods ........... 16 2.2.2 Salmonella subtyping methods during epidemiologic investigations ............. 17 2.3.2.1 Phenotypic methods ..................................................................................... 18 2.2.2.1.1 Serotyping ................................................................................................. 18 2.2.2.1.2 Phage typing .............................................................................................. 18 2.2.2.1.3 Multilocus enzyme electrophoresis (MLEE)............................................. 19 2.2.2.2 Genotypic methods ....................................................................................... 19 2.2.2.2.1 DNA-fragment-pattern-based methods ..................................................... 20 2.2.2.2.1.1 Pulsed-Field Gel Electrophoresis (PFGE) .............................................. 20 2.2.2.2.1.2 Amplified Fragment Length Polymorphism (AFLP) ............................. 22 2.2.2.2.1.3 Multiple Loci Variable number tandem repeat Analysis (MLVA) ........ 23 2.2.2.2.2 DNA-sequence-based methods ................................................................. 24 2.2.2.2.2.1 Multilocus Sequence Typing (MLST) ................................................... 24 2.2.2.2.2.2 Multi-Virulence-Locus Sequence Typing (MVLST) ............................. 26 2.2.2.2.2.3 Single Nucleotide Polymorphism (SNP) analysis .................................. 27
2.3 Clustered Regularly Interspaced Palindromic Repeat (CRISPR) .............................. 28 2.3.1 CRISPR in Salmonella .................................................................................... 30
2.4 Conclusions ................................................................................................................ 30 2.5 References .................................................................................................................. 31
Chapter 3 Novel virulence gene and CRISPR multilocus sequence typing scheme for
subtyping the major serovars of Salmonella enterica subspecies enterica ...................... 46
3.1 Abstract ...................................................................................................................... 47
v
3.2 Introduction ................................................................................................................ 48 3.3 Materials and methods ............................................................................................... 52 3.4 Results ........................................................................................................................ 55 3.5 Discussion .................................................................................................................. 74 3.6 Acknowledgements .................................................................................................... 78 3.7 References .................................................................................................................. 79
Chapter 4 Characterization of clinical, poultry and environmental Salmonella Enteritidis
isolates using multilocus sequence typing based on virulence genes and CRISPRs ....... 86
4.1 Abstract ...................................................................................................................... 87 4.2 Introduction ................................................................................................................ 88 4.3 Materials and methods ............................................................................................... 91 4.4 Results ........................................................................................................................ 93 4.5 Discussion .................................................................................................................. 104 4.6 Acknowledgements .................................................................................................... 108 4.7 References .................................................................................................................. 109
Chapter 5 Conclusions and future research .............................................................................. 115
5.1 Conclusions ................................................................................................................ 115 5.2 Future research ........................................................................................................... 117
APPENDIX Supplemental materials………………………………………………………... 121
vi
LIST OF FIGURES
Figure 2.1 Model for the three-phase evolution of pathogenicity in Salmonella enterica
subspecies enterica. The phylogenetic tree is not drawn to scale (7). ............................ 6
Figure 2.2 Schematic view of the two CRISPR systems in Salmonella Typhimurium LT2.
.......................................................................................................................................... 29
Figure 3.1. Schematic view of the two CRISPR systems in Salmonella Typhimurium
LT2. .................................................................................................................................. 72
Figure 3.2. (a) Cluster diagram based on only fimH and sseL. (b) Cluster diagram based
on fimH, sseL and CRISPRs (combined allele of CRISPR1 and CRISPR2). .................. 73
Figure 4.1. Potential routes of transmission of S. Enteritidis contamination throughout
the egg food system. ......................................................................................................... 98
Figure 4.2. Schematic view of the two CRISPR systems in Salmonella Enteritidis strain
P125109. .......................................................................................................................... 99
Figure 4.3. Frequency of the five predominant sequence types (E ST1, 3, 4, 8 and 10) in
clinical, poultry and environmental isolates. .................................................................... 100
Figure 4.4. Cluster diagram based on only fimH and sseL for all 27 sequence types. ............ 101
Figure 4.5. Cluster diagram based on virulence genes and CRISPRs for all 27 sequence
types. ................................................................................................................................ 102
Figure 4.6. Graphic representation of spacer arrangements in CRISPR1 and CRISPR2 of
the 27 S. Enteritidis sequence types. ................................................................................ 103
Figure S1. Graphic representation of spacer arrangements in CRISPR1 and CRISPR2. ....... 124
vii
LIST OF TABLES
Table 2.1 Top ten most frequently reported serovars from human sources in 2005 ................ 10
Table 2.2 Top ten most frequently reported serovars from human sources in 2006 ................ 10
Table 3.1. Top nine most frequently reported serovars from human sources in 2005
which were analyzed in the present study ........................................................................ 60
Table 3.2. Outbreak information, PFGE profile and MLST results for the 171 isolates
analyzed in the present study ........................................................................................... 61
Table 3.3. Size, function and nucleotide location of the four markers targeted in the
present study .................................................................................................................... 65
Table 3.4. Primers used to amplify and sequence the four MLST markers ............................ 66
Table 3.5. Number of isolates, allelic types and sequence types in each serovar ................... 67
Table 3.6. Allelic polymorphisms and nucleotide substitutions in the nucleotide
sequences of fimH and sseL ............................................................................................. 68
Table 3.7. Analysis of CRISPR repeat sequences .................................................................. 69
Table 3.8. Analysis of CRISPR spacers in different serovars................................................. 70
Table 3.9. Comparison of epidemiologic concordance1 between PFGE and MLST based
on virulence genes and CRISPRs for the selected strains analyzed in the present
study ................................................................................................................................. 71
Table 4.1. Sources, sample types and isolation information for the 167 S. Enteritidis
isolates analyzed in the present study .............................................................................. 96
Table 4.2. Primers used to amplify and sequence the four MLST markers ............................ 97
Table S1. Primers used to amplify and sequence other virulence genes ................................ 121
Table S2. Source, isolate information and MLST results for the 167 isolates analyzed in
the present study ............................................................................................................... 125
viii
LIST OF ABBREVIATIONS AND DEFINITIONS
ADL Animal Diagnostic Lab
AFLP Amplified Fragment Length Polymorphism
bp Base Pair
C Cytosine
CDC Centers for Disease Control and Prevention
°C Degree Celsius
Clone† A group of isolates deriving from a common ancestor as part of a direct
chain of replication and transmission from host to host or from the
environment to host.
CRISPR Clustered Regularly Interspaced Short Palindromic Repeats
D Discriminatory Power
DNA Deoxyribonucleic Acid
dNTP Deoxyribonucleotide Triphosphate
DR Direct Repeat
E Epidemiological Concordance
EC Epidemic Clone
ml milliliter
G Guanine
MLEE Multi-Locus Enzyme Electrophoresis
MLST Multilocus Sequence Typing
MLVA Multiple-Locus Variable-number tandem repeat Analysis
MVLST Multi-Virulence-Locus Sequence Typing
NCBI National Center for Biotechnology Information
ix
PCR Polymerase Chain Reaction
PFGE Pulsed-Field Gel Electrophoresis
PEQAP Pennsylvania Egg Quality Assurance Program
RNA Ribonucleic Acid
rRNA Ribosomal Ribonucleic Acid
SNP Single Nucleotide Polymorphism
Strain† Isolate(s) that exhibit distinct phenotypic and/or genotypic characteristics
from other isolates of the same species
ST Sequence Type
T Thymine
USDA United States Department of Agriculture
μl microliter
WGS Whole Genome Shotgun
† Clone and strain were defined previously by Struelens et al. (101).
x
ACKNOWLEDGEMENTS
I thank my parents, Zijian Liu and Guixiang Liu, who support and encourage me to study
in the US. I am also grateful for the support of my sister, Fenni Liu.
I would like to give my sincere thanks my advisors, Dr. Stephen J. Knabel and Dr.
Edward G. Dudley. I learned from them not only how to do research but also how to lead my life.
I feel so grateful for the working experience with them. I also thank my committee members, Dr.
Rodolphe Barrangou, and Dr. Bhushan M. Jayarao for their guidance and encouragement.
Additionally, I thank Dr. Kariyawasam, Dr. Gerner-Smidt and Dr. Ribot for their help with the
research.
I thank my labmates, Jia Wen, Mei Lok, Gabari, Michelle, Carrie, and Mat for their help
and encouragement. I also want to give special thanks to Dr. Bindhu Verghese for her guidance
and help with my research. Furthermore, I want to thank all the faculty, graduate students and
staff in the Department of Food Science for their support.
At last, I thank USDA and the Department of Food Science for supporting my research.
1
Chapter 1
Statement of the problem
Salmonella is one of the most common foodborne bacteria worldwide. In the United
States alone, there were approximately 1.4 million cases of salmonellosis each year since 1996,
which resulted in a heavy burden on public health and the economy. In order to develop effective
intervention strategies to control salmonellosis during outbreaks, it is critical to rapidly and
accurately track the farm-to-fork spread of Salmonella. Molecular subtyping methods are
powerful tools for investigating the transmission of Salmonella by characterizing specific
outbreak clones. Serotyping has been one of the major subtyping methods employed during
outbreaks to provide base line information about the serovar involved. There are approximately
2,500 different serovars of Salmonella; however, the top ten serovars caused approximately 60%
of all outbreak cases. Each of those top serovars is known to cause numerous outbreaks, each of
which is typically caused by a specific outbreak clone. Therefore, molecular subtyping methods,
which are generally more discriminatory than serotyping, are needed to further distinguish
different strains of a particular serovar. Pulsed-field gel electrophoresis (PFGE) is currently
CDC’s ―gold standard‖ approach for subtyping Salmonella. However, PFGE sometimes lacks
discriminatory power and epidemiologic concordance for typing clonal serovars, such as S.
Enteritidis and S. Montevideo. Many studies have been conducted to develop alternative
subtyping methods, one of which is multi-locus sequence typing (MLST). Previous MLST
schemes for Salmonella focused mainly on discriminatory power; however, none of the previous
MLST studies examined the epidemiologic concordance of the MLST schemes or attempted to
distinguish strains within highly clonal Salmonella serovars, such as S. Enteritidis and S.
2
Montevideo. Moreover, for S. Enteritidis, our knowledge of their epidemiology is hindered due
to its clonal nature. Therefore, the main purpose of the present study was to enhance the
molecular epidemiology of Salmonella by developing an MLST scheme that has both high
discriminatory power and high epidemiologic concordance for subtyping the major serovars of
Salmonella.
3
Chapter 2
Literature review
2.1 Salmonellosis
Salmonella infections (salmonellosis) include three forms of disease: gastroenteritis,
bacteremia and typhoid fever. After ingestion of Salmonella into the gastrointestinal system,
gastroenteritis can develop, which is characterized by symptoms such as abdominal pain, nausea,
vomiting and diarrhea. More severe manifestations of salmonellosis, such as bacteremia and
typhoid fever can develop after the invasion of Salmonella into the bloodstream. Common
symptoms of bacteremia are fever, focal infections, sepsis and meningitis. Typhoid fever is a
deadly systemic infection for humans caused by S. Typhi.
The incidence of typhoid fever has declined in the U.S. with approximately 400 cases
annually (33). On the other hand, infections due to nontyphoidal Salmonella (mainly
gastroenteritis) have increased dramatically during the last 3 to 4 decades (29, 53). The increased
number of infections from nontyphoidal Salmonella may result from modern intensified farming
and food production methods and global trade. Increased spread of Salmonella may also be
promoted by the acquisition of genes for antibiotic resistance (102), and in the case of S.
Enteritidis, genes permitting colonization of chicken ovaries (49).
Globally, it is estimated that there are 93.8 million cases of gastroenteritis due to
Salmonella annually, out of which 80.3 million (86%) cases are foodborne (76). In the United
States, salmonellosis is the leading cause of foodborne bacterial disease, with approximately 1.4
million human cases each year, resulting in 17,000 hospitalizations, 585 deaths (28,116) and a
cost of 2.6 billion dollars due to loss of work, medical care and loss of life (112). Therefore, it is
4
imperative to study the origins, transmission and epidemiology of this pathogen in order to
control and prevent diseases in the future.
2.1.1 Salmonella
Salmonella is one of the most well-known and frequent foodborne bacterial pathogens throughout
the world (76). Salmonella is a genus of rod-shaped, gram negative, non-spore forming,
facultative anaerobic and motile bacteria belonging to the family Enterobacteriaceae.
2.1.2 Salmonella taxonomy and serotyping
The genus Salmonella is comprised of two species: S. enterica and S. bongori. The
species S. bongori is rarely associated with human disease. The species S. enterica has six
subspecies: enterica, salamae, arizonae, diarizonae, houtenae and indica (63, 107). S. enterica
subspecies enterica is responsible for 99% of the human cases of salmonellosis, so it is of greatest
clinical importance (2).
Salmonella subspecies are further differentiated based on serotyping. Serotyping
distinguishes Salmonella immunologically based upon O antigens (lipopolysaccharide) and H
antigens (peritrichous flagella). There are more than 2,500 recognized S. enterica serovars, each
with a unique combination of O and H antigens (54). Prior to 2000, serovars were sometimes
used as species names (16). For example, the original S. typhimurium is now referred to as S.
enterica subspecies enterica serovar Typhimurium or simply S. Typhimurium. The latter
nomenclature is used more commonly in publications and public health surveillance programs
such as those administrated by the Centers for Disease Control and Prevention (CDC).
5
2.1.3 Evolution of pathogenicity
S. enterica subspecies enterica was proposed to evolve in 3 main steps (Fig. 2.1) (7). The
first step involved acquisition of Salmonella pathogenicity island 1 (SPI1) which contributed to
the divergence of Salmonella from E. coli and other related organisms. SPI1 is a 40 kb DNA
region present in both S. enterica and S. bongori (78). It encodes a type III secretion system
(T3SS) required for the intestinal phase of infection and promotes inflammation, the invasion of
intestinal epithelial cells, and secretion of intestinal fluid (117).
The second step of evolution was hypothesized to be the acquisition of a second
pathogenicity island SPI2 in the species S. enterica but not in S. bongori (Fig. 2.1) (7). SPI2
encodes another T3SS and various effector proteins that are required for survival and replication
inside host cells during systemic infection (86, 97). For example, one of the many SPI2 effector
proteins, SseL, is involved in macrophage killing, thus promoting survival inside the host (95).
Due to the presence of SPI2, S. enterica has increased capacity for systemic spread and is thus
more virulent than S. bongori, which do not contain SPI2.
Finally, the host range of S. enterica subspecies enterica expanded to warm-blooded
animals, including humans (Fig. 2.1) (7). In contrast, the other five S. enterica subspecies and S.
bongori are mainly associated with cold-blooded animals. The expansion of host range to warm-
blooded animals requires that bacteria recognize the new hosts for the first step of infection.
Recognition and attachment to the host involves adherence and colonization factors called
adhesins. For example, fimbrial adhesin encoded by the gene fimH allows Salmonella to
recognize and adhere to different receptors on host cells (66, 99). Genetic changes of this gene
by point mutation or recombination might allow the subspecies enterica to recognize new
receptors in new hosts, thus helping to expand its host range. After recognition and attachment,
other processes allowing the subspecies enterica to infect warm blooded animals may include the
ability to survive the immune system and proliferating inside host cells (7). It is not clear which
6
genetic changes accounted for these processes during adaptation to new hosts because adaptation
to a new animal host is a complex process that probably involves a large number of genes.
In summary, acquisition of SPI1 separated the genus Salmonella from other related
organisms like E. coli. Then, acquisition of SPI2 separated the genus Salmonella into two distinct
lineages, S. bongori and S. enterica. Finally, the lineage of S. enterica branched into several
distinct phylogenetic groups. This latter phase of evolution was characterized by host range
expansion of the subspecies enterica to warm-blooded animals, including humans. Through all
these evolutionary steps, Salmonella enterica subspecies enterica (hereafter referred to as
Salmonella) became a highly successful human and animal pathogen.
Figure 2.1 Model for the three-phase evolution of pathogenicity in Salmonella enterica
subspecies enterica. The phylogenetic tree is not drawn to scale (7).
2.1.4 Salmonella reservoirs
Salmonella is mostly transmitted through the fecal-oral route. Salmonellosis occurs when
humans consume foods or water contaminated by animal and human feces containing Salmonella
during food-handling or harvesting. Therefore, foods serve as the main transmission vector for
7
Salmonella, which include animal foods that are not thoroughly cooked and contaminated
uncooked vegetables and fruits (116).
Generally speaking, transmission of Salmonella starts from its reservoirs, which are
defined as any person, animal, plant, soil or substance (or combination of these) in which a
microorganism normally lives and grows (67). Salmonella serovars have adapted to live in a
variety of hosts. Many wild animals, such as gorillas (10), rhinoceros (68), lizards (88), reptiles
and snakes (9) harbor Salmonella. More importantly, food animals including chickens, turkeys,
cattle, swine and sheep have also been found to frequently carry Salmonella.
Different serovars have different reservoirs and modes of pathogenesis. For example, S.
Typhi, which causes the deadly disease typhoid fever, is a strict human pathogen. Some other
serovars, such as S. Gallinarum in chickens, S. Choleraesuis in swine and S. Dublin in cattle, are
known to be associated mainly with one animal, but rarely cause disease in humans. In contrast,
other serovars like S. Typhimurium have adapted to a broad host range, including wild and
domestic animals and humans. Moreover, different animals have different predominant serovars
associated with them. Predominant serovars associated with poultry, cattle and swine will be
reviewed here in brief because those animals are the primary vectors for transmitting Salmonella
to humans and are the main focus of this study.
The most prevalent and important reservoirs for Salmonella are poultry (23). The most
common poultry-associated serovars, Enteritidis in eggs and Typhimurium in poultry, accounted
for 33.3 % of the total human foodborne diseases in the U.S. (20). The top 5 most common
serovars associated with broilers are Kentucky, Heidelberg, Enteritidis, Typhimurium and I 4, [5],
12: i: - (113). They represent 81% of all Salmonella isolates from broilers. Similarly, serovars
Hadar, Heidelberg, Reading, Schwarzengrund, and Saintpaul account for 68% of all Salmonella
isolates from turkeys (113).
8
Cattle are also frequently found to harbor Salmonella. They can carry many different
serovars of Salmonella, with Montevideo, Anatum, Muenster, Newport, Mbandanka the most
common serovars that account for 47 % of Salmonella isolates from cattle (114).
As for swine, another important reservoir for Salmonella, the 5 most frequent serovars
are Derby, Typhimurium, Infantis, Anatum and Saintpaul. These 5 serovars comprise 60% of all
isolates from swine (114).
It is noteworthy that most of these serovars found predominantly in food animals are the
same serovars that are frequently associated with human diseases. Given this fact, it is of great
importance to control and monitor levels of the most common serovars in animals and
subsequently prevent their transmission to humans.
2.1.5 Salmonella association with foods
Another important vehicle for transmitting Salmonella to humans is produce. Salmonella
can cycle through the food chain and the environment in soil, water, manure, and insects.
Therefore, contamination of produce can occur by various ways throughout the food system.
Like predominant serovars in animals, there are also predominant produce-associated serovars,
which include Enteritidis, Newport, Poona, Typhimurium, Braenderup, Javiana, Montevideo and
Muenchen (60). The overlap between serovars most commonly associated with animals and
those associated with produce suggests contamination of produce during growing or harvesting
processes directly or indirectly by animals containing Salmonella. Moreover, evidence is
accumulating that enteric bacteria have the ability to grow and persist on and in plants, such as
tomatoes, radish sprouts, bean sprouts, barley, and lettuce (15, 47, 62).
Contamination and persistence of Salmonella on produce promote the transmission of
this pathogen to humans. Salmonella outbreaks associated with fresh produce have increased in
the U.S in recent years (98). Many kinds of produce have been linked to Salmonella outbreaks,
9
such as tomatoes, sprouts, melons, cantaloupe, lettuce, peppers and mangos (98). Produce causes
the highest number of human diseases and second highest number of outbreaks among various
food vehicles in the U.S. (3). For example, the largest Salmonella outbreak to date occurred in
2008 and was caused by consumption of Jalapeño and Serrano peppers that were contaminated
with S. Saintpaul (22).
Besides foods of animal origin and produce, there has been an increase in Salmonella
outbreaks caused by new food vehicles, such as salami, peanut butter, veggie booty, pot pies, and
dry cereals. For instance, in 2010, Italian-style salami and its ingredients (red and black peppers
containing S. Montevideo) caused a multistate outbreak which infected 252 people from 44 states
(27). As a result, approximately 1,378,754 pounds of Italian sausage products were recalled by
Daniele International, Inc. (27). Another recent outbreak caused by a new food vehicle is the
2008-2009 peanut butter outbreak, which infected 714 people from 46 states and caused 6 deaths
(24). As a result, more than 2,100 peanut-containing products were recalled by over 200
companies.
Outbreaks due to those new food vehicles were not expected because they are more or
less processed foods which do not possess conditions that permit the growth of Salmonella. For
example, peanut butter is a dry food with an aw below the minimum level for growth (0.94).
Moreover, Salmonella can be inhibited or killed by heat, acid, high salt concentration, etc. during
food manufacturing processes (38). Persistence of Salmonella in processed foods might be due to
1) high levels of Salmonella in food ingredients; 2) inadequate sanitary practices; 3) and the
ubiquity of Salmonella in animals, produce and the environment.
2.1.6 Most common Salmonella serovars associated with human illnesses
Although there are over 2,500 Salmonella serovars, only a handful of Salmonella
serovars caused most human illnesses (Tables 2.1 and 2.2) (20, 21).
10
Table 2.1 Top ten most frequently reported serovars from human sources in 2005
Rank Serovar No. of laboratory-confirmed cases % of total cases
1 Typhimurium 6982 19.3
2 Enteritidis 6730 18.6
3 Newport 3295 9.1
4 Heidelberg 1903 5.3
5 Javiana 1324 3.7
6 I 4, [5], 12: i :- 822 2.3
7 Montevideo 809 2.2
8 Muenchen 733 2
9 Saintpaul 683 1.9
10 Braenderup 603 1.7
total 66
Laboratory-confirmed cases include both outbreak cases and sporadic cases.
Source: 2005 Salmonella annual review (20).
Table 2.2 Top ten most frequently reported serovars from human sources in 2006
Rank Serovar No. of laboratory-confirmed cases % of total cases
1 Typhimurium 6872 16.9
2 Enteritidis 6740 16.6
3 Newport 3373 8.3
4 Heidelberg 1495 3.7
5 Javiana 1433 3.5
6 I 4, [5], 12: i :- 1200 3.0
7 Montevideo 1061 2.6
8 Muenchen 753 1.9
9 Oranienburg 719 1.8
10 Mississippi 604 1.5
total 60
Laboratory-confirmed cases include both outbreak cases and sporadic cases.
Source: 2006 Salmonella annual review (21).
11
Compared to all the other serovars of Salmonella, S. Typhimurium caused the highest
number of human illnesses and was associated with a broad range of foods (Table 2.3). As
mentioned before, S. Typhimurium has adapted to various hosts, including birds, amphibians, and
all food animals, especially poultry, cattle and swine. Not only can S. Typhimurium reside in so
many animals, but it can also be found in them at high frequency (114). The ubiquity and
relatively high numbers of S. Typhimurium might explain why it has caused so many outbreaks
via so many kinds of foods (Table 2.3).
The second most common serovar is S. Enteritidis, which caused nearly as many human
cases as S. Typhimurium (Tables 2.1 and 2.2). The major food vehicles for S. Enteritidis are shell
eggs, as 80% of the S. Enteritidis outbreaks were egg-associated (89). S. Enteritidis contaminates
eggs either through horizontal transmission, by which eggs are externally contaminated by feces
containing S. Enteritidis (36), or by vertical transmission, where the inside of the eggs is
contaminated by infected ovaries before the laying of the egg (50, 87). Vertical transmission is
believed to be the more important route because eggs contaminated by vertical transmission
produce a new generation of infected broilers or layers after hatching (50, 57, 79). In order to
control S. Enteritidis in poultry, one of the interventions employed in the U.S. is egg quality
assurance programs on farms. These voluntary programs involve acquisition of S. Enteritidis free
chicks, control of pests (including rodents and flies), use of S. Enteritidis-free feeds, and routine
microbiologic testing for S. Enteritidis in the farm environment (14).
The third most commonly reported serovar causing salmonellosis is S. Newport (Tables
2.1 and 2.2). S. Newport can be detected in many food animals, but is most frequently isolated
from cattle (113). S. Newport has been implicated in many outbreaks via a variety of food
vehicles, such as beef, chicken, pork, tomatoes, cantaloupes, melons, avocadoes and guacamole
12
(23). In 2010, S. Newport caused a multistate outbreak due to contaminated alfalfa sprouts, in
which 35 people became ill (26). Cases of illness caused by S. Newport have increased in recent
years, which might be due to the emerging multidrug-resistant S. Newport isolates (19).
The fourth most common serovar is S. Heidelberg (Tables 2.1 and 2.2). It is often
isolated from commercial broilers and ground chicken (113). As a result, poultry and eggs have
been identified as the major food vehicles for this serovar (32). The largest outbreak caused by S.
Heidelberg occurred in 2007, when 802 people became infected via contaminated hummus (Table
2.3).
Following S. Heidelberg, S. Javiana caused the fifth most human infections (Tables 2.1
and 2.2). Unlike other serovars, S. Javiana is rarely isolated from poultry, cattle or swine (113).
The major reservoirs for S. Javiana were considered to be amphibians, as direct contact with
amphibians has been associated with outbreaks. Amphibian feces-contaminated tomatoes were
identified to be the main food vehicles for S. Javiana (34). For example, tomatoes were identified
to be the food source of S. Javiana for a multistate outbreak in 2002, which resulted in 159 cases
(Table 2.3).
The sixth most common serovar I 4, [5], 12: i :- , a variant of serovar S. Typhimurium, is
antigenically similar to S. Typhimurium, but lacks the second-phase
flagella antigens (39). It is
also one of the most commonly identified serovar in broilers and ground chicken (113). I 4, [5],
12: i :- contaminated pot pies caused a multistate outbreak in 2007 (Table 2.3).
S. Montevideo is the next most commonly reported serovar. S. Montevideo is frequently
isolated from cattle and ground beef (113). Food vehicles of S. Montevideo include beef, turkey,
pork and sprouts (22). The most recent outbreak caused by S. Montevideo occurred in 2010 due
to contaminated Italian-style meats (27).
The eighth most common serovar is S. Muenchen. S. Muenchen can be detected in swine,
cattle, chicken etc. It has been associated with outbreaks due to multiple food vehicles, such as
13
chicken, sprouts, tomato, and cantaloupe (22). In 1999, a multistate outbreak was caused by S.
Muenchen in orange juice, which infected 398 people.
S. Saintpaul ranks as the ninth most common serovar in 2005, but dropped to eleventh in
2006 (20, 21). However, its ranking might have risen higher since then, because it caused the
largest Salmonella outbreak in 2008 due to contaminated peppers. S. Saintpaul is frequently
isolated from swine and has caused outbreaks due to foods like sprouts, tomatoes, mangoes,
orange juice, turkey etc.
The importance of the above top serovars is reflected by the high number of
salmonellosis cases they cause. Their success as human pathogens might be largely due to
adaptation to food animals. For example, 4 of the top 8 serovars are frequently found in poultry,
namely Typhimurium, Enteritidis, Heidelberg and I 4, [5], 12: i :-. Two other serovars, Newport
and Montevideo, are mainly found in cattle.
14
Table 2.3 Salmonella outbreaks caused by the top 8 serovars in the United States from 1998- 2010
Year Serovar Ill Hospitalizations Deaths Food vehicle
2008 Typhimurium 530 116 8 peanut butter
2001 Typhimurium 404 0 4 unidentified
2006 Typhimurium 199 39 0 deli meat
2006 Typhimurium 192 24 0 tomato
2005 Typhimurium 162 0 sauces; fajita
2006 Typhimurium 161 7 0 chicken
1998 Typhimurium 134 10 0 multiple foods
2002 Typhimurium 132 0 0 unidentified
2002 Typhimurium 116 4 0 milk
1999 Typhimurium 112 3 0 clover sprouts
2002 Typhimurium 107 6 0 milk
2007 Typhimurium 87 8 0 Veggie Booty
2007 Typhimurium 76 4 0 lettuce; spinach
2003 Typhimurium 67 2 0 eggs
2007 Typhimurium 66 3 0 pork
2003 Typhimurium 59 2 0 beef
2005 Typhimurium 57 8 0 cake
2003 Typhimurium 56 11 0 ground beef
1998 Typhimurium 50 1 0 smoked fish
2003 Typhimurium 50 7 0 queso fresco
2002 Enteritidis 700 3 0 salsa
2005 Enteritidis 304 56 1 turkey
1999 Enteritidis 256 0 0 ice cream
2001 Enteritidis 231 34 0 egg-based sauce
2002 Enteritidis 196 24 0 cake
2005 Enteritidis 126 15 0 cantaloupe
2006 Enteritidis 113 23 0 oil; chicken
2001 Enteritidis 113 0 0 eggs
2000 Enteritidis 106 14 0 macaroni cheese
2007 Enteritidis 106 14 0 chicken
2003 Enteritidis 104 12 0 crab cakes
2001 Enteritidis 92 7 0 eggs
2002 Enteritidis 90 2 0 beef; pork
2000 Enteritidis 88 orange juice
1999 Enteritidis 82 3 0 honeydew melon
2002 Newport 510 tomato
2006 Newport 115 8 0 tomato
2004 Newport 100 5 0 milk
2004 Newport 97 lettuce
2000 Newport 96 6 0 pico de gallo
1999 Newport 79 mango
2006 Newport 77 2 0 turkey
2003 Newport 68 13 2 honeydew melon
2007 Newport 67 5 0 pork
2007 Newport 65 11 0 tomato
2004 Newport 49 8 0 turkey and gravy
15
2002 Newport 47 12 1 ground beef
2007 Newport 46 tomato; avocado
2007 Heidelberg 802 29 0 hummus
2003 Heidelberg 517 chicken
2002 Heidelberg 239 22 0 beef
1998 Heidelberg 200 4 0 cake
2002 Heidelberg 104 22 macaroni cheese
2007 Heidelberg 79 mashed potato
2004 Heidelberg 78 2 0 turkey
2005 Heidelberg 75 5 0 sandwich; vanilla cake
2003 Heidelberg 65 14 0 Swiss cheese
2003 Heidelberg 57 7 0 eggs; pancakes
2000 Heidelberg 56 3 0 macaroni salad
1999 Heidelberg 41 chicken
2003 Javiana 227 9 0 fajita, chicken
2002 Javiana 159 3 0 tomato
2004 Javiana 60 1 0 beans
2000 Javiana 44 8 0 bread; chicken
2007 I 4,[5],12:i :- 401 108 3 pot pie
2010 Montevideo 252 Italian-style meats
2006 Montevideo 72 19 0 sandwich, beef
2002 Montevideo 55 6 0 beef
1999 Muenchen 398 orange juice
1999 Muenchen 61 6 0 alfalfa sprouts
2003 Muenchen 58 15 cantaloupe
2002 Muenchen 57 3 0 pasta salad
2005 Saintpaul ;
Typhimurium
157 orange juice
2008 Saintpaul 1442 286 2 peppers
2009 Saintpaul 235 alfalfa sprouts
Source: CDC foodborne outbreak database (23).
2.2 Subtyping of Salmonella
In order to control Salmonella outbreaks, it is important to trace back the sources and
identify the routes by which Salmonella are transmitted to foods. However, trace-back
investigation of outbreaks can be hindered due to the complexity of the food chain and the
limitations of traditional epidemiologic investigations. The limitations of traditional
epidemiologic investigations include 1) Only a limited number of cases are reported; 2) People
tend not to recall the foods that were eaten before disease onset; 3) Cases are often spread out in
16
time and space; and 4) Investigations can be hindered if the food source is not listed on the
investigation questionnaire (60).
Based on the reasons above, another trace-back method called subtyping is carried out
along with traditional epidemiologic investigations. Subtyping characterizes bacteria at the strain
level (101). By characterizing the outbreak-related strains and separating them from non-related
strains, subtyping can play an essential role in investigating Salmonella outbreaks.
Besides tracking pathogens in epidemiologic investigations, the other use of subtyping
methods is to study the population structure, evolution and diversity of bacteria on a long-term
scale. For example, one subtyping method called multilocus enzyme electrophoresis (MLEE) has
been used to study the genetic diversity of Salmonella populations (8). Studies like this can
provide insight into the evolutionary history and emergence of Salmonella serovars. However,
the focus of this review is on the short-term epidemiologic applications of subtyping methods.
2.2.1 Important definitions and performance criteria of subtyping methods
Before considering the epidemiology of Salmonella, it is important to first clarify the
definitions for outbreak, epidemic, strain, epidemic clone (EC), and outbreak clone (OC) used
frequently in epidemiologic studies. These definitions were previously compiled by Chen and
Knabel (30). Outbreak is an acute appearance of a cluster of an illness that occurs in numbers in
excess of what is expected for that time and place. Epidemic is defined as one or more outbreaks
that spread widely over a long period of time. Strain is defined as isolates that have distinct
phenotypic and genotypic characteristics from other isolates from the same species. Epidemic
clone is a strain or group of strains descended asexually from a single ancestral cell (source strain)
that is involved in one epidemic, and can often include several outbreaks. Outbreak clone is a
strain or group of strains descended asexually from a single ancestral cell (source strain) that is
involved in one outbreak (30).
17
To evaluate and compare different subtyping schemes, there are several performance
criteria, which include typeability, reproducibility, discriminatory power and epidemiologic
concordance. Typeability is the capability of a method to generate an interpretable result for each
strain typed. For example, strains that do not have plasmids cannot be typed by plasmid profiles.
Reproducibility is the ability of a subtyping method to generate the same result each time the
sample is tested. Discriminatory power is the ability of a subtyping method to differentiate
between unrelated epidemic or outbreak clones. Epidemiologic concordance is the capacity of a
typing method to correctly cluster epidemic and outbreak clones, and separate them from clones
that are not epidemiologically related (101). Many studies of subtyping methods focused on the
discriminatory power of the subtyping system. On the other hand, few studies have examined the
epidemiologic concordance of a particular subtyping method. The reason for the lack of studies
examining epidemiologic concordance might be that most studies did not utilize well-defined
strains from multiple outbreaks.
The choice of strain collection is critical when developing and evaluating a new
subtyping system for outbreak investigations. As mentioned before, an ideal strain collection
should include well-defined strains from multiple common-source outbreaks in order to access
both discriminatory power and epidemiologic concordance. A good subtyping system should
separate strains from different outbreaks, but not separate strains within the same
outbreak/outbreak clone.
2.2.2 Salmonella subtyping methods during epidemiologic investigations
Subtyping methods can be either phenotypic or genotypic approaches. Phenotypic
methods include screening for antibiotic resistance, bacteriophage susceptibility and surface
antigens, such as the H and O antigens. Genotypic methods differentiate strains based on
differences in genome sequence and/or structure. Major phenotypic and genotypic subtyping
18
methods available for Salmonella will be briefly discussed here with the primary focus on
genotypic methods.
2.3.2.1 Phenotypic methods
Before the advent of genotypic methods, many phenotypic methods were widely used for
typing Salmonella strains. Common phenotypic methods for Salmonella include serotyping,
phage typing and MLEE. In general, although phenotypic methods provide useful information
about the strains, they often lack enough discriminatory power.
2.2.2.1.1 Serotyping
As mentioned in the taxonomy section, serotyping distinguishes Salmonella based on
immunological classification of the H and O antigens (54). Serotyping is one of the most
important phenotypic methods for Salmonella, which provides baseline information before other
typing methods can be carried out to further separate strains in a particular serovar. Serotyping is
very useful because the serovar name often points to the specific reservoir and mode of
pathogenesis. However, serotyping alone is not suit for molecular epidemiology, because
individual serovars are responsible for multiple outbreaks (20, 21). As a result, other subtyping
methods with more resolution need to be carried out after serotyping.
2.2.2.1.2 Phage typing
Phage typing utilizes the selective capacity of individual bacteriophage to infect bacterial
cells. During phage typing, a panel of bacteriophages is used to infect bacteria and phage types
are assigned according to the patterns of lysis. Phage typing has been shown to be a good
19
indicator for pandemic clones of Salmonella. For instance, S. Enteritidis phage type (PT) 4 is the
most common PT in Europe, while PT8 is the most common PT in the U.S. Another example is
S. Typhimurium definitive type 104 (DT104), which is typically resistant to a number of
antibiotics and has had a major impact on global health (106). However, phage typing sometimes
suffers from low typeability in that many strains are resistant to all typing phages (1). Moreover,
it requires maintenance of the typing phage stocks and specially trained personnel (45).
2.2.2.1.3 Multilocus enzyme electrophoresis (MLEE)
MLEE differentiates strains based on the relative electrophoretic mobility of cellular
enzymes. The variation in amino acid sequences of the enzymes from different strains results in
differences in electrostatic charges. This leads to different migrations of the enzymes in an
electric field. By comparing the electrophoretic profiles, genetic relatedness of strains can then
be determined. MLEE has been carried out to analyze the population structure of Salmonella
serovars and the relatedness of strains within a serovar (8). Population studies by MLEE
subtyping revealed that while many serovars have similar electrophoretic types (ETs) that form a
single cluster, other serovars like S. Newport have divergent ETs clustered distantly in MLEE
trees. Using MLEE to determine phylogenetic relationships of bacteria is generally accepted.
However, MLEE has been replaced by a more reproducible and portable method called
multilocus sequence typing (MLST), which looks directly at DNA sequences of several genes
(75). MLST will be introduced later as one of the genotypic methods.
2.2.2.2 Genotypic methods
Genotypic methods target genetic differences between different strains of bacteria.
Generally speaking, genotypic methods have better reproducibility and increased discriminatory
20
power than phenotypic methods. Because of these advantages, genotypic methods are often
carried out after serotyping during Salmonella outbreak investigations. Two categories of
genotypic methods, DNA-fragment-pattern-based methods and DNA-sequence-based methods,
will be discussed.
2.2.2.2.1 DNA-fragment-pattern-based methods
Three DNA-fragment-pattern-based subtyping methods have been extensively studied for
subtyping Salmonella, which are pulsed-field gel electrophoresis (PFGE), amplified fragment
length polymorphism (AFLP) and multiple loci variable number tandem repeat analysis (MLVA).
2.2.2.2.1.1 Pulsed-Field Gel Electrophoresis (PFGE)
PFGE is currently the ―gold standard‖ method for subtyping Salmonella and is used by
public health surveillance systems such as the PulseNet program of CDC. During PFGE
procedures, bacterial cells are first immobilized in agarose plugs to avoid mechanical shearing of
the long genomic DNA. Cells in agarose plugs are then lysed and genomic DNA is digested by a
rare-cutting restriction endonuclease. Next, agarose plugs containing digested genomic DNA are
put into wells of an agarose gel. The agarose gel is then subjected to an electric field whose
orientation is periodically changing. This pulsed electrical field can resolve large DNA fragments
that could not be separated by a constant unidirectional electrical field. The standardized PFGE
protocol of Salmonella uses two restriction endonucleases XbaI and BlnI in separate reactions
(40).
PFGE has been used in detection, investigation and control of numerous outbreaks and is
generally very successful (51). The main advantage of PFGE is its comparatively high
discriminatory power for subtyping most serovars of Salmonella. However, PFGE lacks
21
discriminatory power for clonal serovars like Enteritidis (25, 120) and Montevideo (27), or clonal
phage types like S. Typhimurium DT104 (51). This is reflected by low PFGE pattern diversity
for those serovars and clonal phage types in the PulseNet database (51). In the cases of such low
discriminatory power, outbreak clones cannot be separated from sporadic isolates and other non-
outbreak related isolates, which can hinder epidemiologic detection and investigation. For
example, during the recent Italian-style meat outbreak, the outbreak clone of S. Montevideo had
the most common PFGE pattern in PulseNet database, which made it difficult to detect the
outbreak (27).
Besides low discriminatory power for clonal serovars, another limitation of PFGE is the
ambiguous interpretation of banding patterns. Banding patterns can change due to insertions,
deletions and point mutations. For instance, a single nucleotide mutation might cause up to 3-
fragment changes in the PFGE banding pattern. Because of this difficulty, interpretation of PFGE
banding patterns has been proposed to follow several guidelines: 1) strains showing no fragment
differences with the outbreak strain are part of the outbreak; 2) strains showing 1 fragment
difference with the outbreak strain are probably part of the outbreak; 3) strains showing 2-3
fragment differences with the outbreak strain are possibly part of the outbreak; 4) strains showing
more than 3-fragment differences with the outbreak strain are not part of the outbreak (105).
More recommendations for interpretation of PFGE patterns have been published recently. The
recommendations include taking into account the quality of the PFGE gel, the diversity of the
organism and the temporal and geographical information during analysis of PFGE patterns (40).
Although those suggestions helped standardize the interpretation of PFGE patterns, these
recommendations are still not completely objective.
Another drawback of PFGE is low reproducibility if the standardized protocol is not
strictly followed. As a result, subsequent comparison of PFGE banding patterns cannot be carried
out, especially when comparing PFGE patterns between different laboratories. To overcome this
limitation, PulseNet implemented an extensive quality assurance system (51). This system
22
requires laboratories to obtain PFGE gel preparation and gel analysis certification and participate
in the annual proficiency testing program. All these steps help ensure comparability and
reproducibility, but at the same time it requires personnel specially trained by the quality
assurance system.
To sum up, although it is the current ―gold standard‖ subtyping method, PFGE suffers
from several drawbacks which limit its performance for subtyping Salmonella.
2.2.2.2.1.2 Amplified Fragment Length Polymorphism (AFLP)
AFLP is a method that employs both restriction digestion and polymerase chain reaction
(PCR) techniques. In AFLP, genomic DNA is digested with one or more restriction enzymes.
The ends of the digested DNA fragments are then ligated to adaptors that are complementary to
the restriction sites. The digested and ligated DNA fragments are then selectively amplified using
PCR primers targeting the adaptor sequences. PCR primers typically contain one to three
additional nucleotides on their 3’-end to reduce the number of amplified fragments to a
manageable number. PCR products are then subjected to electrophoresis and characteristic
banding patterns are then produced.
AFLP is a relatively simple and fast approach. The discriminatory power of AFLP is
equal to that of PFGE for subtyping S. Typhimurium (73, 103), but higher than that of PFGE for
subtyping S. Enteritidis (52) and other serovars (109). However, its discriminatory power has
been reported to be insufficient to separate all epidemiologically unrelated S. Typhimurium
strains (92).
Like PFGE, the reproducibility of AFLP among different laboratories is problematic
since comparing AFLP results among different laboratories is difficult (48). Variability in the
AFLP profile can be generated by minor changes in the amplification conditions. Therefore,
replicates of the sample could be identified as different strains (45). To enhance reproducibility,
23
PCR should be performed under highly stringent conditions (84) and gel electrophoresis should
be standardized.
2.2.2.2.1.3 Multiple Loci Variable number tandem repeat Analysis (MLVA)
MLVA targets tandem repeats of short DNA sequences in bacterial genomes. The
difference in the number of repeated DNA motifs is employed to differentiate strains. In a
MLVA assay, a number of well-selected and characterized loci are amplified by PCR using
primers targeting the flanking regions of the repeated loci. PCR products are then separated and
the number of repeat units at each locus can be measured according to the size of the PCR
products. Differences in the number of repeats in each locus are used to distinguish different
strains.
Since this method is based on PCR, MLVA has the advantage of being easy to perform
and rapid. Moreover, MLVA yields discreet and unambiguous data, reported as the number of
repeat units at each locus. Comparison of MLVA profiles between laboratories can be made with
a simple nomenclature recently proposed (70). The discriminatory power of MLVA was reported
to be higher than PFGE and AFLP for subtyping S. Typhimurium (72, 108) and higher than
PFGE for S. Enteritidis (11, 93). However, in some circumstances, strains that have the same
MLVA type were separated by PFGE profiles (13). This indicates that strains of same MLVA
type might not be closely related.
However, the reproducibility of MLVA is a potential problem. The instability of MLVA
alleles has been observed for subtyping S. Newport and S. Typhimurium (18, 35). Replicates of
the same strains have been shown to have different number of repeat units at a specific locus (35).
The instability of the MLVA loci is probably due to DNA polymerase slippage during genome
replication (110). This instability might make interpretation difficult when strains have slightly
different MLVA types.
24
To conclude, by providing improved discriminatory power and having a short turnaround
time, MLVA can be used as a complementary method to PFGE in epidemiologic investigations of
Salmonella. MLVA has been used successfully along with other subtyping methods in outbreak
investigations to track Salmonella (12, 83, 85). However, MLVA also suffered from some
drawbacks and thus it has not been widely used for this purpose.
2.2.2.2.2 DNA-sequence-based methods
DNA-sequence-based methods differentiate strains by the detection of polymorphic DNA
sequences. Multilocus sequence typing (MLST) and single nucleotide polymorphism (SNP)
analysis are both DNA-sequence-based methods and will be briefly reviewed here.
2.2.2.2.2.1 Multilocus Sequence Typing (MLST)
MLST discriminates among bacterial strains by comparing nucleotide sequences of
several DNA loci in bacteria chromosomes. For each locus in the MLST scheme, every new
allele is assigned a unique number in order of discovery and is designated an allelic type. The
collective allelic types make up the allelic profile or sequence type, which may also be assigned a
unique and arbitrary number. For example, in the MLST database (www.mlst.net) based on the
seven loci: aroC, dnaN, hemD, hisD, purE, sucA, and thrA, one of the strains in the database has
an allelic profile of (1, 1, 2, 1, 1, 1, 9) for each of the seven genes, and was assigned sequence
type 3 (80). The collective allelic types and sequence types are compared among bacterial strains
and then cluster analysis can be carried out.
Compared to PFGE, MLST is a less labor-intensive method and involves common
techniques including primer design, PCR amplification and DNA sequencing. Furthermore,
DNA sequence represents discreet, unambiguous, highly informative, highly portable and
25
reproducible data. Many MLST data sets are available over the internet (www.mlst.net) so that a
uniform nomenclature is ensured and comparison of results among laboratories can be conducted
rapidly. The application of MLST is promoted due to the increased speed and reduced cost of
nucleotide sequencing and improved internet database and tools (74). These advantages make
MLST an attractive subtyping approach.
MLST schemes originally target housekeeping genes, which are genes required for
fundamental metabolic functions and are found within all members of a given species (75). For
example, 7 housekeeping genes were targeted in the first MLST scheme for Neisseria
meningitidis (75). Housekeeping genes are excellent genetic markers for studying the population
structure, long-term evolution and diversity of bacteria. A good overview of Salmonella diversity
and evolution is provided by the internet-based MLST data. Based upon MLST data, Salmonella,
especially S. enterica subspecies enterica, is highly clonal (69). Moreover, the data suggest that
many serovars including Typhimurium, Enteritidis, Newport and Saintpaul may have more than
one origin (69). However, MLST schemes based on housekeeping genes for typing Salmonella
usually have much lower discriminatory power than that of PFGE (43, 61, 109). The results of
those studies suggested that housekeeping genes do not provide sufficient resolution to
distinguish closely related strains. Therefore, MLST schemes based on housekeeping genes are
not suitable for outbreak investigations.
To conclude, MLST possesses many attractive advantages. It is an excellent tool for
global phylogenetic studies. However, housekeeping genes selected in previous MLST studies
lacked sequence variation and thus were ineffective for subtyping Salmonella for epidemiologic
purposes. To track strains of this important pathogen during outbreaks, genetic markers that give
sufficient DNA sequence variations need to be identified.
26
2.2.2.2.2.2 Multi-Virulence-Locus Sequence Typing (MVLST)
Besides housekeeping genes, virulence genes which are responsible for pathogenesis,
have been selected as genetic markers for MLST schemes. MLST schemes that only target
virulence genes have been referred to as multi-virulence-locus sequence typing (MVLST) (31,
119). Unlike housekeeping genes, virulence genes are commonly under positive selection (41).
As a result, DNA sequences of virulence genes tend to be more variable than housekeeping genes
and thus are able to provide increased discrimination. It is also speculated that virulence genes
can provide high epidemiologic concordance because they are responsible for causing diseases
and thus outbreaks. For example, six virulence genes were targeted in an MVLST scheme for
subtyping Listeria monocytogenes, which showed very high discriminatory power (0.99) and
perfect epidemiologic concordance (1.0) (31).
No MVLST scheme has yet been developed for subtyping Salmonella. However, MLST
based on both virulence genes and housekeeping genes has been published for typing Salmonella
enterica subspecies enterica serovars, which targeted flagellin genes fliC and fljB along with two
housekeeping genes, gyrB and atpD (104). This study included several strains from all
subspecies and 22 of the more prevalent Salmonella enterica subspecies enterica serovars
attempting to develop a DNA-based assay for serotype identification. However, the use of this
MLST scheme to further characterize strains under serovar level was not tested. Another MLST
based on both virulence genes and housekeeping genes has been developed for subtyping S.
Typhimurium and showed high discriminatory power (0.98), which was slightly higher than that
of PFGE (0.96) (46). In that MLST scheme, three virulence genes were included together with
the 16S rRNA gene and three housekeeping genes. One of the virulence genes in that MLST
scheme is hilA which regulates transcription of invasion proteins (4). The other two virulence
genes, pefB and fimH, encode different fimbriae and both mediate adherence to host cells (6, 66).
Although this MLST scheme seems to have adequate discriminatory power for subtyping S.
27
Typhimurium, its capacity to discriminate strains from more clonal serovars such as S. Enteritidis
has not yet been tested. Currently, there is no published MLST study for differentiating strains
within S. Enteritidis. In the SNP database of NCBI (National Center for Biotechnology
Information), two strains of S. Enteritidis were compared side by side to examine their SNPs (56).
Nearly all virulence genes were identical between the two, suggesting that MVLST might not be
discriminatory enough for differentiating strains of S. Enteritidis.
In summary, although MVLST has higher discriminatory power than MLST using
housekeeping genes, it may not provide enough discrimination for clonal serovars like Enteritidis.
In order to develop an MLST scheme for outbreak investigations, additional genetic markers with
even higher sequence variability need to be identified.
2.2.2.2.2.3 Single Nucleotide Polymorphism (SNP) analysis
SNP analysis differentiates strains by nucleotide substitutions at specific sites in the
bacterial genome. SNP analysis often involves three steps: 1) Select SNP sites that are variable to
provide discrimination among strains; 2) Determine the nucleotide bases at the selected sites of
different strains; and 3) Compare the SNPs among strains. Selection of the SNP sites is often
based on previous knowledge of specific polymorphic genes (42, 71) or comparative genomic
studies (118). To determine the nucleotide base (adenine, guanine, cytosine, and thymine) at a
defined SNP site, multiple methods can be used, such as pyrosequencing or realtime PCR (82, 91,
111).
Because SNP analysis targets SNPs in the bacterial genome, it has the potential to be
more rapid and cost efficient than MLST. However, there are very few SNP analysis studies for
subtyping Salmonella. SNP analysis targeting genes associated with quinolone resistance has
been used to study the antibiotic resistance of Salmonella (42, 71). Another SNP analysis study
targeted SNPs in flagella antigens in order to develop a SNP typing method to replace serotyping
28
(82). No SNP typing methods have been developed for differentiating Salmonella strains for
outbreak investigations. The reason might be that the SNP loci of Salmonella that could provide
the desirable discrimination have not been identified.
In conclusion, although SNP analysis has the potential to be rapid, cost efficient and
high-throughput, the lack of information about SNP sites suitable for subtyping Salmonella make
it difficult to develop a SNP typing protocol for epidemiologic purposes.
2.3 Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)
Since virulence genes alone might not provide enough discrimination for subtyping
clonal Salmonella serovars, additional genetic elements that are evolving faster than virulence
genes are needed. One of the fastest evolving genetic elements in bacteria genomes are CRISPRs
(Clustered Regularly Interspaced Palindromic Repeat) (100). CRISPRs are regions of direct
repeats (DRs) and spacers in the chromosomes of archaea and bacteria, including S. enterica (Fig.
2.2) (65, 100). DRs are 21-47 bp long, separated by spacers of similar size (Fig. 2.2). Sequences
of DRs are generally conserved, except the repeat at one end of the CRISPR is not totally
conserved, and is thus called a degenerate direct repeat (Fig. 2.2). On the other hand, sequences
of spacers are quite variable from each other. It was recently demonstrated that CRISPR spacers
are derived from phages or plasmids, which when inserted into the CRISPR of a bacterial cell
help protect that cell from subsequent infection by those same phages and plasmids (5). CRISPR
is generally flanked at one end by a common leader sequence of 200-350 bp, which is believed to
act as a promoter to transcribe CRISPR into small RNAs (77). Immediately upstream from the
CRISPR there are CRISPR-associated (Cas) proteins that carry functional domains of nucleases,
helicases, polymerases and polynucleotide-binding proteins (58). Some Cas proteins can
recognize foreign DNA invading the bacteria, and then integrate a new repeat-spacer unit into
CRISPR at the leader end. Therefore, when the same exogenous nucleic acid invades next time,
29
the CRISPR transcribed crRNAs (CRISPR RNAs) can recognize the foreign nucleic acid and
lead the Cas proteins to degrade these invading nucleic acid (17, 59, 65). In this way, CRISPR
along with Cas proteins can block foreign sequences, such as sequences of phages and plasmids.
Figure 2.2 Schematic view of the two CRISPR systems in Salmonella Typhimurium LT2.
Direct repeats and spacers are represented by black diamonds and white rectangles, respectively.
The degenerate direct repeats are represented by white diamonds. Numbers of direct repeats and
spacers are represented by the numbers of diamonds and white rectangles, respectively. L stands
for leader sequence. cas genes are in grey while other core flanking genes (ygcF, iap and ptps)
are in white. The graph is not drawn to scale.
As a bacterial immune system against phages and plasmids, CRISPRs evolve rapidly and
adaptively (115). As mentioned before, new spacers could be added when foreign DNA invades
the bacteria. Besides addition of new spacers, deletion of spacers is also frequently observed (37,
90). However, the mechanism of deletion of spacers is not clear. The addition of new spacers
and deletion of one or several spacers make CRISPR one of the most variable DNA loci in
bacteria and form a high degree of polymorphism among strains (90).
CRISPRs have been used for subtyping Mycobacterium tuberculosis, this subtyping
method is called Spacer oligotyping or spoligotyping (55). In this method, PCR is carried out
using primers designed according to the sequence of the DR so that each spacer can be amplified.
The PCR products are then hybridized to a membrane containing probes for specific spacers. The
hybridization patterns showing the presence or absence of spacers are then compared among
strains. Spoligotyping is now the standard method for subtyping M. tuberculosis for outbreak
CRISPR1
CRISPR2
30
investigations. It has also been used in subtyping Corynebacterium diphtheria (81). Other than
spoligotyping, CRISPR sequence analysis has also been used for other bacteria, such as Yersinia
pestis (90), Streptococcus (64), and Campylobacter jejuni (96). As for Salmonella, although
CRISPRs have the potential to be excellent markers for separating Salmonella strains, they have
not been widely used for subtyping purposes.
2.3.1 CRISPR in Salmonella
CRISPR can be found in multiple numbers in bacteria. Two CRISPR loci are found in all
Salmonella serovars in the CRISPRs database (http://crispr.u-psud.fr/crispr/). CRISPR direct
repeats in Salmonella are 27-31 bp long. Salmonella CRISPRs have great polymorphism even
among strains belonging to the same serovar. Therefore, CRISPRs might serve as good markers
for subtyping Salmonella during epidemiologic investigations.
2.4 Conclusions
Salmonella is the leading cause of foodborne bacterial disease in the U.S. Most human
illnesses are caused by a handful of serovars, such as Typhimurium, Enteritidis, Newport,
Heidelberg, I 4, [5], 12; i: -, Montevideo, Muenchen and Saintpaul. Salmonella can reside in
many wild and domestic animals and can spread from numerous reservoirs to contaminate
numerous kinds of foods, which makes it especially challenging to track this pathogen during
outbreaks. Therefore, to reduce outbreaks caused by the most common serovars of Salmonella, it
is critical to employ a subtyping method that can accurately identify its sources and pathways of
transmission. Many subtyping methods have been developed for differentiating Salmonella
strains, such as PFGE, AFLP and MLVA. Each method has its own advantages and drawbacks.
PFGE is currently the ―gold standard‖ method for outbreak investigations. However, PFGE
31
produces ambiguous data that are hard to interpret and more importantly PFGE often lacks
discriminatory power for subtyping clonal serovars such as Enteritidis. In contrast, MLST
generates highly informative and discreet data consisting of nucleotide sequences that can be
easily interpreted and rapidly compared on internet databases. Previous MLST schemes targeting
housekeeping genes were not very successful largely due to low discriminatory power associated
with conserved housekeeping genes. Unlike housekeeping genes, virulence genes can provide
important information about the pathogenesis of strains and improve the discriminatory power of
MLST. However, the discriminatory power of virulence genes may still not be enough for
subtyping clonal serovars of Salmonella. CRISPRs are one of the fastest evolving genetic
elements that could be implemented in an MLST scheme to provide increased discrimination. In
order to develop an MLST scheme for outbreak investigation, virulence genes and CRISPRs were
targeted in the present study to subtype the top 10 serovars of Salmonella. This MLST scheme
was speculated to provide high discriminatory power and epidemiologic concordance for
subtyping Salmonella for epidemiologic purposes.
2.5 References
1. Amavisit, P., P. F. Markham, D. Lightfoot, K. G. Whithear, and G. F. Browning. 2001.
Molecular epidemiology of Salmonella Heidelberg in an equine hospital. Vet. Microbiol.
80:85-98.
2. Anjum, M. F., C. Marooney, M. Fookes, S. Baker, G. Dougan, A. Ivens, and M. J.
Woodward. 2005. Identification of Core and Variable Components of the Salmonella
entericas subspecies I genome by microarray. Infect. Immun. 73:7894-7905.
3. Anonymous. 2006. Outbreak alert! Closing the gaps in our federal food safety net.
http://www.cspinet.org/new/pdf/outbreakalert2004.pdf
32
4. Bajaj, V., R. L. Lucas, C. Hwang, and C. A. Lee. 1996. Co-ordinate regulation of
Salmonella Typhimurium invasion genes by environmental and regulatory factors is mediated
by control of hilA expression. Mol. Microbiol.. 22:703-714.
5. Barrangou, R., C. Fremaux, H. Deveau, M. Richards, P. Boyaval, S. Moineau, D. A.
Romero, and P. Horvath. 2007. CRISPR provides acquired resistance against viruses in
prokaryotes. Science. 315:1709-1712.
6. Baumler, A., R. Tsolis, F. Bowe, J. Kusters, S. Hoffmann, and F. Heffron. 1996. The pef
fimbrial operon of Salmonella Typhimurium mediates adhesion to murine small intestine and
is necessary for fluid accumulation in the infant mouse. Infect. Immun. 64:61-68.
7. Baumler, A. J., R. M. Tsolis, T. A. Ficht, and L. G. Adams. 1998. Evolution of host
adaptation in Salmonella enterica. Infect. Immun. 66:4579-4587.
8. Beltran, P., J. M. Musser, R. Helmuth, J. J. Farmer, W. M. Frerichs, I. K. Wachsmuth,
K. Ferris, A. C. McWhorter, J. G. Wells, and A. Cravioto. 1988. Toward a population
genetic analysis of Salmonella: genetic diversity and relationships among strains of serotypes
S. Choleraesuis, S. Derby, S. Dublin, S. Enteritidis, S. Heidelberg, S. Infantis, S. Newport, and
S. Typhimurium. Proc. Natl. Acad. Sci. U.S.A. 85:7753-7757.
9. Bemis, D. A., L. M. Grupka, S. Liamthong, D. W. Fooland, J. M. Sykesiv, and E. C.
Ramsay. 2007. Clonal relatedness of Salmonella isolates associated with captive and wild-
caught rattlesnakes. Vet. Microbiol. 300-307.
10. Benirschke, K., and F. D. Adams. 1980. Gorilla diseases and causes of death. J. Reprod.
Fertil. 28:139-148.
11. Beranek, A., C. Mikula, P. Rabold, D. Arnhold, C. Berghold, I. Lederer, F. Allerberger,
and C. Kornschober. 2009. Multiple-locus variable-number tandem repeat analysis for
subtyping of Salmonella enterica subsp. enterica serovar Enteritidis. Int. J. Med. Microbiol.
299:43-51.
33
12. Best, E. L., M. D. Hampton, S. Ethelberg, E. Liebana, F. A. Clifton-Hadley, and E. J.
Threlfall. 2009. Drug-resistant Salmonella Typhimurium DT 120: use of PFGE and MLVA
in a putative international outbreak investigation. Microb. Drug Resist.15:133-138.
13. Boxrud, D., K. Pederson-Gulrud, J. Wotton, C. Medus, E. Lyszkowicz, J. Besser, and J.
M. Bartkus. 2007. Comparison of multiple-locus variable-number tandem repeat analysis,
pulsed-field gel electrophoresis, and phage typing for subtype analysis of Salmonella enterica
serotype Enteritidis. J. Clin. Microbiol. 45:536-543.
14. Braden, C. R. 2006. Salmonella enterica serotype Enteritidis and eggs: a national epidemic
in the United States. Clin. Infect. Dis. 43:512-517.
15. Brandl, M. T. 2006. Fitness of human enteric pathogens on plants and implications for food
safety. Annu. Rev. Phytopathol. 44:367-392.
16. Brenner, F. W., R. G. Villar, F. J. Angulo, R. Tauxe, and B. Swaminathan. 2000.
Salmonella Nomenclature. J. Clin. Microbiol. 38:2465-2467.
17. Brouns, S. J. J., M. M. Jore, M. Lundgren, E. R. Westra, R. J. H. Slijkhuis, A. P. L.
Snijders, M. J. Dickman, K. S. Makarova, E. V. Koonin, and J. van der Oost. 2008.
Small CRISPR RNAs guide antiviral defense in prokaryotes. Science. 321:960-964.
18. Call, D., L. Orfe, M. Davis, S. Lafrentz, and M. Kang. 2008. Impact of compounding error
on strategies for subtyping pathogenic bacteria. Foodborne Pathog. Dis.5:505-516.
19. CDC. 2002. Outbreak of multidrug-resistant Salmonella Newport--United States, January-
April 2002. JAMA. 288:951-953.
20. CDC. 2006. Salmonella Annual Summary 2005.
http://www.cdc.gov/ncidod/dbmd/phlisdata/salmtab/2006/SalmonellaAnnualSummary2006.p
df
34
21. CDC. 2007. Salmonella Annual Summary 2006.
http://www.cdc.gov/ncidod/dbmd/phlisdata/salmtab/2006/SalmonellaAnnualSummar
y2006.pdf
22. CDC. 2008. Outbreak of Salmonella serotype Saintpaul infections associated with multiple
raw produce items --- United States, 2008. MMWR Morb Mortal Wkly Rep. 57:929-934.
23. CDC. 2008. OutbreakNet Foodborne Outbreak Online Database.
http://wwwn.cdc.gov/foodborneoutbreaks/
24. CDC. 2009. Multistate outbreak of Salmonella infections associated with peanut butter and
peanut butter--containing products --- United States, 2008--2009. MMWR Morb Mortal
Wkly Rep. 58:1-6.
25. CDC. 2010. Investigation update: multistate outbreak of human Salmonella Enteritidis
infections associated with shell eggs. http://www.cdc.gov/salmonella/enteritidis/
26. CDC. 2010. Investigation announcement: multistate outbreak of human Salmonella Newport
infections linked to raw alfalfa sprouts. http://www.cdc.gov/salmonella/newport/index.html
27. CDC. 2010. Investigation update: multistate outbreak of human Salmonella Montevideo
infections. http://www.cdc.gov/salmonella/montevideo/index.html
28. CDC. 2010. Preliminary FoodNet data on the incidence of infection with pathogens
transmitted commonly through food --- 10 States, 2009. MMWR Morb Mortal Wkly Rep.
58:333-337
29. Chalker, R. B., and M. J. Blaser. 1988. A review of human salmonellosis: III. magnitude of
Salmonella infection in the United States. Rev. Infect. Dis. 10:111-124.
30. Chen, Y., and S. Knabel. 2008. Strain typing, p. 203-239. In D. Liu (ed.), Handbook of
Listeria monocytogenes.
35
31. Chen, Y., W. Zhang, and S. J. Knabel. 2007. Multi-virulence-locus sequence typing
identifies single nucleotide polymorphisms which differentiate epidemic clones and outbreak
strains of Listeria monocytogenes. J. Clin. Microbiol. 45:835-846.
32. Chittick, P., A. Sulka, R. V. Tauxe, and A. M. Fry. 2006. A summary of national reports of
foodborne outbreaks of Salmonella Heidelberg infections in the United States. J. Food Prot.
69:1150-1153.
33. Christie, A. B. 1987. Infectious diseases: epidemiologic and clinical practice. Churchill
Livingstone, New York.
34. Clarkson, L. S., M. Tobin-D’angelo, C. Shuler, S. Hanna, J. Benson, and A. Voetsch.
2010. Sporadic Salmonella enterica serotype Javiana infections in Georgia and Tennessee: a
hypothesis-generating study. Epidemiol. Infect. 138:340-346.
35. Davis, M. A., K. N. K. Baker, D. R. Call, L. D. Warnick, Y. Soyer, M. Wiedmann, Y.
Grohn, P. L. McDonough, D. D. Hancock, and T. E. Besser. 2009. Multilocus variable-
number tandem-repeat method for typing Salmonella enterica serovar Newport. J. Clin.
Microbiol. 47:1934-1938.
36. De Reu, K., K. Grijspeerdt, W. Messens, M. Heyndrickx, M. Uyttendaele, J. Debevere,
and L. Herman. 2006. Eggshell factors influencing eggshell penetration and whole egg
contamination by different bacteria, including Salmonella Enteritidis. Int. J. Food Microbiol.
112:253-260.
37. Deveau, H., R. Barrangou, J. E. Garneau, J. Labonte, C. Fremaux, P. Boyaval, D. A.
Romero, P. Horvath, and S. Moineau. 2007. Phage response to CRISPR-encoded resistance
in Streptococcus thermophilus. J. Bacteriol. 190:1390-1400.
38. Doyle, M. P., and L. R. Beuchat eds. 2007. Food microbiology: fundamentals and frontiers.
ASM Press, Washington, DC.
36
39. Echeita, M. A., S. Herrera, and M. A. Usera. 2001. Atypical, fljB-negative Salmonella
enterica subsp. enterica strain of serovar 4,5,12:i:- appears to be a monophasic variant of
serovar Typhimurium. J. Clin. Microbiol. 39:2981-2983.
40. Efrain, M. R., M. A. Fair, R. Gautom, D. N. Cameron, S. B. Hunter, B. Swaminathan,
and T. J. Barrett. 2006. Standardization of pulsed-field gel electrophoresis protocols for the
subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne
Pathog. Dis. 3:59-67.
41. Endo, T., K. Ikeo, and T. Gojobori. 1996. Large-scale search for genes on which positive
selection may operate. Mol. Biol. Evol. 13:685-690.
42. Esaki, H., K. Noda, N. Otsuki, A. Kojima, T. Asai, Y. Tamura, and T. Takahashi. 2004.
Rapid detection of quinolone-resistant Salmonella by real time SNP genotyping. J. Microbiol.
Methods. 58:131-134.
43. Fakhr, M. K., L. K. Nolan, and C. M. Logue. 2005. Multilocus sequence typing lacks the
discriminatory ability of pulsed-field gel electrophoresis for typing Salmonella enterica
serovar Typhimurium. J. Clin. Microbiol. 43:2215-2219.
44. Fakhr, M. K., J. S. Sherwood, J. Thorsness, and C. M. Logue. 2006. Molecular
characterization and antibiotic resistance profiling of Salmonella isolated from retail turkey
meat products. Foodborne Pathog. Dis. 3:366-374.
45. Foley, S. L., S. Zhao, and R. D. Walker. 2007. Comparison of molecular typing methods
for the differentiation of Salmonella foodborne pathogens. Foodborne Pathog. Dis. 4:253-276.
46. Foley, S. L., D. G. White, P. F. McDermott, R. D. Walker, B. Rhodes, P. J. Fedorka-
Cray, S. Simjee, and S. Zhao. 2006. Comparison of subtyping methods for differentiating
Salmonella enterica serovar Typhimurium isolates obtained from food animal sources. J. Clin.
Microbiol. 44:3569-3577.
47. Franz, E., and A. H. C. van Bruggen. 2008. Ecology of E. coli O157:H7 and Salmonella
enterica in the primary vegetable production chain. Crit. Rev. Microbiol. 34:143-161.
37
48. Fry, N. K., B. Afshar, P. Visca, D. Jonas, J. Duncan, E. Nebuloso, A. Underwood, and T.
G. Harrison. 2005. Assessment of fluorescent amplified fragment length polymorphism
analysis for epidemiological genotyping of Legionella pneumophila serogroup 1. Clin.
Microbiol. Infect. 11:704-712.
49. Gantois, I., R. Ducatelle, F. Pasmans, F. Haesebrouck, R. Gast, T. J. Humphrey, and F.
Van Immerseel. 2009. Mechanisms of egg contamination by Salmonella Enteritidis. FEMS
Microbiology Reviews. 33:718-738.
50. Gast, R. K., and C. W. Beard. 1990. Production of Salmonella Enteritidis-contaminated
eggs by experimentally infected hens. Avian Dis. 34:438-446.
51. Gerner-Smidt, P., K. Hise, J. Kincaid, S. Hunter, S. Rolando, E. Hyytiä-Trees, E. M.
Ribot, B. Swaminathan, and Pulsenet Taskforce. 2006. Pulsenet USA: a five-year update.
Foodborne Pathog. Dis. 3:9-19.
52. Giammanco, G. M., C. Mammina, C. Romani, I. Luzzi, A. M. Dionisi, and A. Nastasi.
2007. Evaluation of a modified single-enzyme amplified fragment length polymorphism (SE-
AFLP) technique for subtyping Salmonella enterica serotype Enteritidis. Res. Microbiol.
158:10-17.
53. Gomez, T. M., Y. Motarjemi, S. Miyagawa, F. K. Käferstein, and K. Stöhr. 1997.
Foodborne salmonellosis. World Health Stat Q. 50:81-89.
54. Grimont, P. A. D., and F. X. Weill. 2007. Antigenic formulae of the Salmonella serovars,
9th ed. WHO Collaborating Centre for Reference and Research on Salmonella. Institute
Pasteur, Paris, France.
55. Groenen, P. M., A. E. Bunschoten, D. Soolingen, and J. D. Errtbden. 1993. Nature of
DNA polymorphism in the direct repeat cluster of Mycobacterium tuberculosis; application
for strain differentiation by a novel typing method. Mol. Microbiol.10:1057-1065.
38
56. Guard, J. 2010. Evolutionary trends in two strains of Salmonella enterica subsp. I serovar
Enteritidis PT13a that vary in virulence potential.
http://www.ncbi.nlm.nih.gov/genomes/static/Salmonella_SNPS.html
57. Guard-Petter, J. 2001. The chicken, the egg and Salmonella Enteritidis. Environ. Microbiol.
3:421-430.
58. Haft, D. H., J. Selengut, E. F. Mongodin, and K. Nelson. 2005. A Guild of 45 CRISPR-
associated (Cas) protein families and multiple CRISPR/cas subtypes exist in prokaryotic
genomes. PLoS Comput Biol. 1:e60.
59. Hale, C. R., P. Zhao, S. Olson, M. O. Duff, B. R. Graveley, L. Wells, R. M. Terns, and M.
P. Terns. 2009. RNA-guided RNA cleavage by a CRISPR RNA-cas protein complex. Cell.
139:945-956
60. Hanning B. Irene, J.D. Nutt, Steven C. Ricke. 2009. Salmonellosis outbreaks in the United
States due to fresh produce: sources and potential intervention measures. Foodborne Pathog.
Dis. 6:635-648.
61. Harbottle, H., D. G. White, P. F. McDermott, R. D. Walker, and S. Zhao. 2006.
Comparison of multilocus sequence typing, pulsed-field gel electrophoresis, and
antimicrobial susceptibility typing for characterization of Salmonella enterica serotype
Newport isolates. J. Clin. Microbiol.
62. Harris, L. J., J. N. Farber, L. R. Beuchat, M. E. Parish, T. V. Suslow, E. H. Garrett, and
F. F. Busta. 2003. Outbreaks associated with fresh produce: incidence, growth, and survival
of pathogens in fresh and fresh-cut produce. Compr Rev Food Sci Food Saf. 2:78-141.
63. Heyndrickx, M., F. Pasmans, R. Ducatelle, A. Decostere, and F. Haesebrouck. 2005.
Recent changes in Salmonella nomenclature: The need for clarification. Vet. J. 170:275-277.
64. Hoe, N., K. Nakashima, D. Grigsby, X. Pan, S. J. Dou, S. Naidich, M. Garcia, E. Kahn,
D. Bergmire-Sweat, and J. M. Musser. 1999. Rapid molecular genetic subtyping of
serotype M1 group A Streptococcus strains. Emerg Infect Dis. 5:254-263.
39
65. Horvath, P., and R. Barrangou. 2010. CRISPR/cas, the immune system of bacteria and
archaea. Science. 327:167-170.
66. Humphries, A. D., S. M. Townsend, R. A. Kingsley, T. L. Nicholson, R. M. Tsolis, and A.
J. Baumler. 2001. Role of fimbriae as antigens and intestinal colonization factors of
Salmonella serovars. FEMS Microbiology Letters. 201:121-125.
67. Jay, J. M. 2005. Modern Food Microbiology. Springer Science, New York.
68. kenny, D. E. 1999. Salmonella spp. survey of captive rhinoceroses in U.S. zoological
institutions and private ranches. J. Zoo Wildl. Med. 30:383-388.
69. Lan, R., P. R. Reeves, and S. Octavia. 2009. Population structure, origins and evolution of
major Salmonella enterica clones. Infect. Genet. Evol. 9:996-1005
70. Larsson, J. T., M. Torpdahl, R. F. Petersen, G. Sørensen, B. A. Lindstedt, and E. M.
Nielsen. 2009. Development of a new nomenclature for Salmonella Typhimurium multilocus
variable number of tandem repeats analysis (MLVA). Euro Surveill. 14:19174-19174.
71. Levy, D. D., B. Sharma, and T. A. Cebula. 2004. Single-nucleotide polymorphism mutation
spectra and resistance to quinolones in Salmonella enterica serovar Enteritidis with a mutator
phenotype. Antimicrob. Agents Chemother. 48:2355-2363.
72. Lindstedt, B., E. Heir, E. Gjernes, and G. Kapperud. 2003. DNA fingerprinting of
Salmonella enterica subsp. enterica serovar Typhimurium with emphasis on phage type
DT104 based on variable number of tandem repeat loci. J. Clin. Microbiol. 41:1469-1479.
73. Lindstedt, B., E. Heir, T. Vardund, and G. Kapperud. 2000. Fluorescent amplified-
fragment length polymorphism genotyping of Salmonella enterica subsp. enterica serovars
and comparison with pulsed-field gel electrophoresis typing. J. Clin. Microbiol. 38:1623-
1627.
74. Maiden, M. C. J. 2006. Multilocus sequence typing of bacteria. Annu. Rev. Microbiol.
60:561-588.
40
75. Maiden, M. C. J., J. A. Bygraves, E. Feil, G. Morelli, J. E. Russell, R. Urwin, Q. Zhang,
J. Zhou, K. Zurth, D. A. Caugant, I. M. Feavers, M. Achtman, and B. G. Spratt. 1998.
Multilocus sequence typing: a portable approach to the identification of clones within
populations of pathogenic microorganisms. Proc. Natl. Acad. Sci. U.S.A. 95:3140-3145.
76. Majowicz, S., J. Musto, E. Scallan, F. Angulo, M. Kirk, S. O’Brien, T. Jones, A. Fazil,
and R. Hoekstra. 2010. Food safety: The global burden of nontyphoidal Salmonella
gastroenteritis. Clin. Infect. Dis.50:882-889.
77. Marraffini, L. A., and E. J. Sontheimer. Self versus non-self discrimination during
CRISPR RNA-directed immunity. Nature. 463:568-571.
78. Mills, D. M., V. Bajaj, and C. A. Lee. 1995. A 40 kb chromosomal fragment encoding
Salmonella typhimurium invasion genes is absent from the corresponding region of the
Escherichia coli K-12 chromosome. Mol. Microbiol. 15:749-v.
79. Miyamoto, T., E. Baba, T. Tanaka, K. Sasai, T. Fukata, and A. Arakawa. 1997.
Salmonella Enteritidis contamination of eggs from hens inoculated by vaginal, cloacal, and
intravenous routes. Avian Dis. 41:296-303.
80. MLST. 2010. http://mlst.ucc.ie/
81. Mokrousov, I., E. Limeschenko, A. Vyazovaya, and O. Narvskaya. 2007.
Corynebacterium diphtheriae spoligotyping based on combined use of two CRISPR loci.
Biotechnol J. 2:901-906.
82. Mortimer, C., T. Peters, S. Gharbia, J. Logan, and C. Arnold. 2004. Towards the
development of a DNA-sequence based approach to serotyping of Salmonella enterica. BMC
Microbiol. 4:31.
83. Much, P., J. Pichler, S. Kasper, H. Lassnig, C. Kornschober, A. Buchner, C. König, and
F. Allerberger. 2009. A foodborne outbreak of Salmonella Enteritidis phage type 6 in
Austria, 2008. Wien. Klin. Wochenschr. 121:132-136.
41
84. Mueller, U. G., and L. L. Wolfenbarger. 2008. AFLP genotyping and fingerprinting.
Trends Ecol Evol. 14:389-394.
85. Munnoch, S., K. Ward, S. Sheridan, G. Fitzsimmons, C. Shadbolt, J. Piispanen, Q.
Wang, T. Ward, T. Worgan, C. Oxenford, J. Musto, J. McAnulty, and D. Durrheim.
2009. A multi-state outbreak of Salmonella Saintpaul in Australia associated with cantaloupe
consumption. Epidemiol Infect. 137:367-374.
86. Ochman, H., F. C. Soncini, F. Solomon, and E. A. Groisman. 1996. Identification of a
pathogenicity island required for Salmonella survival in host cells. Proc. Natl. Acad. Sci.
U.S.A. 93:7800-7804.
87. Okamura, M., Y. Kamijima, T. Miyamoto, H. Tani, K. Sasai, and E. Baba. 2001.
Differences among six Salmonella serovars in abilities to colonize reproductive organs and to
contaminate eggs in laying hens. Avian Dis. 45:61-69.
88. Pasmans, F., A. Martel, F. Boyen, D. Vandekerchove, I. Wybo, F. V. Immerseel, M.
Heyndrickx, J. M. Collard, R. Ducatelle, and F. Haesebrouck. 2005. Characterization of
Salmonella isolates from captive lizards. Vet. Microbiol. 110:285-291.
89. Patrick, M. E., P. M. Adcock, T. M. Gomez, S. F. Altekruse, B. H. Holland, R. V. Tauxe,
and D. L. Swerdlow. 2004. Salmonella Enteritidis infections, United States, 1985-1999.
Emerging Infect. Dis. 10:1-7.
90. Pourcel, C., G. Salvignol, and G. Vergnaud. 2005. CRISPR elements in Yersinia pestis
acquire new repeats by preferential uptake of bacteriophage DNA, and provide additional
tools for evolutionary studies. Microbiology. 151:653-663.
91. Roos, A., P. Dieltjes, R. H. A. M. Vossen, M. R. Daha, and P. de Knijff. 2006. Detection
of three single nucleotide polymorphisms in the gene encoding mannose-binding lectin in a
single pyrosequencing reaction. J. Immunol. Methods. 309:108-114.
42
92. Ross, I. L., and M. W. Heuzenroeder. 2005. Use of AFLP and PFGE to discriminate
between Salmonella enterica serovar Typhimurium DT126 isolates from separate food-
related outbreaks in Australia. Epidemiol. Infect. 133:635-644.
93. Ross, I. L., and M. W. Heuzenroeder. 2009. A comparison of two PCR-based typing
methods with pulsed-field gel electrophoresis in Salmonella enterica serovar Enteritidis. Int. J.
Med. Microbiol. 299:410-420.
94. Rychlik, I., D. Gregorova, and H. Hradecka. 2006. Distribution and function of plasmids
in Salmonella enterica. Vet. Microbiol. 112:1-10.
95. Rytkönen, A., J. Poh, J. Garmendia, C. Boyle, A. Thompson, M. Liu, P. Freemont, J. C.
D. Hinton, and D. W. Holden. 2007. SseL, a Salmonella deubiquitinase required for
macrophage killing and virulence. Proc. Natl. Acad. Sci. U.S.A. 104:3502-3507.
96. Schouls, L. M., S. Reulen, B. Duim, J. A. Wagenaar, R. J. L. Willems, K. E. Dingle, F. M.
Colles, and J. D. A. Van Embden. 2003. Comparative genotyping of Campylobacter jejuni
by amplified fragment length polymorphism, multilocus sequence typing, and short repeat
sequencing: strain diversity, host range, and recombination. J. Clin. Microbiol. 41:15-26.
97. Shea, J. E., M. Hensel, C. Gleeson, and D. W. Holden. 1996. Identification of a virulence
locus encoding a second type III secretion system in Salmonella Typhimurium. Proc. Natl.
Acad. Sci. U.S.A. 93:2593-2597.
98. Sivapalasingam, S., C. R. Friedman, L. Cohen, and Robert V. T. 2004. Fresh produce: a
growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J.
Food Prot. 67:2342-2353.
99. Sokurenko, E. V., H. S. Courtney, D. E. Ohman, P. Klemm, and D. L. Hasty. 1994.
FimH family of type 1 fimbrial adhesins: functional heterogeneity due to minor sequence
variations among fimH genes. J. Bacteriol. 176:748-755.
43
100. Sorek, R., V. Kunin, and P. Hugenholtz. 2008. CRISPR — a widespread system that
provides acquired resistance against phages in bacteria and archaea. Nat. Rev. Microbiol.
6:181-186.
101. Struelens, M. J. 1996. Consensus guidelines for appropriate use and evaluation of microbial
epidemiologic typing systems. Clin. Microbiol. Infect. 2:2-11.
102. Swartz, M. 2002. Human diseases caused by foodborne pathogens of animal origin. Clin.
Infect. Dis. 34:S111-S122.
103. Tamada, Y., Y. Nakaoka, K. Nishimori, A. Doi, T. Kumaki, N. Uemura, K. Tanaka, S.
Makino, T. Sameshima, M. Akiba, M. Nakazawa, and I. Uchida. 2001. Molecular typing
and epidemiological study of Salmonella enterica serotype Typhimurium isolates from cattle
by fluorescent amplified-fragment length polymorphism fingerprinting and pulsed-field gel
electrophoresis. J. Clin. Microbiol. 39:1057-1066.
104. Tankouo-Sandjong, B., A. Sessitsch, E. Liebana, C. Kornschober, F. Allerberger, H.
Hächler, and L. Bodrossy. 2006. MLST-v, multilocus sequence typing based on virulence
genes, for molecular typing of Salmonella enterica subsp. enterica serovars. J. Microbiol.
Methods. 69:23-36.
105. Tenover, F. C., R. D. Arbeit, R. V. Goering, P. A. Mickelsen, B. E. Murray, D. H.
Persing, and B. Swaminathan. 1995. Interpreting chromosomal DNA restriction patterns
produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin
Microbiol. 33:2233-2239.
106. Threlfall, E. J. 2000. Epidemic Salmonella Typhimurium DT 104--a truly international
multiresistant clone. J. Antimicrob. Chemother. 46:7-10.
107. Tindall, B. J., P. A. D. Grimont, G. M. Garrity, and J. P. Euzeby. 2005. Nomenclature
and taxonomy of the genus Salmonella. Int. J. Syst. Evol. Microbiol. 55:521-524.
44
108. Torpdahl, M., G. Sørensen, B. Lindstedt, and E. M. Nielsen. 2007. Tandem repeat
analysis for surveillance of human Salmonella Typhimurium infections. Emerg Infect Dis.
13:388-395.
109. Torpdahl, M., M. N. Skov, D. Sandvang, and D. L. Baggesen. 2005. Genotypic
characterization of Salmonella by multilocus sequence typing, pulsed-field gel
electrophoresis and amplified fragment length polymorphism. J. Microbiol. Methods. 63:173-
184.
110. Torres-Cruz, J., and M. W. van der Woude. 2003. Slipped-strand mispairing can function
as a phase variation mechanism in Escherichia coli. J. Bacteriol. 185:6990-6994.
111. Tyagi, S., D. P. Bratu, and F. R. Kramer. 1998. Multicolor molecular beacons for allele
discrimination. Nat. Biotechnol. 16:49-53.
112. USDA-ERS. 2009. Foodborne illness cost calculator: Salmonella.
http://www.ers.usda.gov/Data/FoodborneIllness/salm_Intro.asp
113. USDA-FSIS. 2006. Serotypes profile of Salmonella isolates from meat and poultry products
January 1998 through December 2005.
http://www.fsis.usda.gov/Science/Serotypes_Profile_Salmonella_Isolates/index.asp
114. USDA-FSIS. 2007. Progress report on Salmonella testing of raw meat and poultry products,
1998-2006. http://www.fsis.usda.gov/science/progress_report_salmonella_testing/index.asp
115. Vale, P. F., and T. J. Little. 2010. CRISPR-mediated phage resistance and the ghost of
coevolution past. Proc. R. Soc. Lond., B, Biol. Sci. 277:2097-2103.
116. Voetsch, A., T. Van Gilder, F. Angulo, M. Farley, S. Shallow, R. Marcus, P. Cieslak, V.
Deneen, and R. Tauxe. 2004. FoodNet estimate of the burden of illness caused by
nontyphoidal Salmonella infections in the United States. Clin. Infect. Dis. 38:S127-S134.
117. Wallis, T. S., and E. E. Galyov. 2000. Molecular basis of Salmonella-induced Enteritis. Mol.
Microbiol. 36:997-1005.
45
118. Zhang, W., W. Qi, T. J. Albert, A. S. Motiwala, D. Alland, E. K. Hyytia-Trees, E. M.
Ribot, P. I. Fields, T. S. Whittam, and B. Swaminathan. 2006. Probing genomic diversity
and evolution of Escherichia coli O157 by single nucleotide polymorphisms . Genome Res.
16:757-767.
119. Zhang, W., B. M. Jayarao, and S. J. Knabel. 2004. Multi-virulence-locus sequence typing
of Listeria monocytogenes. Appl. Environ. Microbiol. 70:913-920.
120. Zheng, J., C. E. Keys, S. Zhao, J. Meng, and E. W. Brown. 2007. Enhanced subtyping
scheme for Salmonella Enteritidis. Emerging Infect. Dis. 13:1932-1935.
46
Chapter 3 Novel virulence gene and CRISPR multilocus sequence typing
scheme for subtyping the major serovars of Salmonella enterica subspecies
enterica
Fenyun Liu1, Rodolphe Barrangou
2, Peter Gerner-Smidt
3, Efrain Ribot
3, Stephen Knabel
1, and
Edward Dudley1*
1Department of Food Science, the Pennsylvania State University, University Park, Pennsylvania
16802;
2 Danisco USA Incorporation, 3329 Agriculture Drive, Madison, Wisconsin 53716
3 Centers for Disease Control and Prevention, Atlanta, Georgia 30333
*Corresponding author. Mailing address: 326 Food Science Building, The Pennsylvania State
University, University Park, PA 16802, US. Phone: 814-867-0439. Email: [email protected]
47
3.1 Abstract
Salmonella enterica subsp. enterica is the leading cause of bacterial foodborne disease in
the United States. Molecular subtyping methods are powerful tools for tracking the farm-to-fork
spread of foodborne pathogens during outbreaks. In order to develop a novel multilocus
sequence typing (MLST) scheme for subtyping the major serovars of S. enterica subspecies
enterica, the virulence genes sseL and fimH and Clustered Regularly Interspaced Short
Palindromic Repeat (CRISPR) regions were sequenced from 171 clinical isolates from serovars
Typhimurium, Enteritidis, Newport, Heidelberg, Javiana, I 4, [5], 12; i: -, Montevideo, Muenchen
and Saintpaul. The MLST scheme using only virulence genes identified epidemic clones, but was
insufficient to separate outbreak clones. However, the addition of CRISPR sequences
dramatically improved discriminatory power of this MLST method by accurately differentiating
individual outbreak clones. Moreover, the present MLST scheme provided better discrimination
of S. Enteritidis strains than PFGE. Cluster analyses also revealed the current MLST scheme is
highly congruent with serotyping. In conclusion, the novel MLST scheme described in the
present study accurately differentiated outbreak clones of the major serovars of Salmonella, and
therefore maybe an excellent method for subtyping this important foodborne pathogen during
outbreak investigations.
48
3.2 Introduction
Salmonella enterica subsp. enterica (Salmonella) is the leading cause of bacterial
foodborne disease in the United States, with approximately 1.4 million human cases each year
since 1996, resulting in an estimated 17,000 hospitalizations, more than 500 deaths (10, 53) and a
cost of 2.6 billion dollars (51). The nine most common serovars, S. Typhimurium, S. Enteritidis,
S. Newport, S. Heidelberg, S. Javiana, S. I 4, [5], 12; i: -, S. Montevideo, S. Muenchen and S.
Saintpaul, were responsible for more than 60% of human illnesses based on the Centers for
Disease Control and Prevention’s (CDC’s) annual summaries of 2005 and 2006 (Table 3.1) (4, 5).
Salmonella has been isolated from a broad range of foods, including raw animal foods (poultry,
eggs, pork, beef, mutton and seafood), produce (sprouts, lettuce, spinach, tomatoes and peppers)
and various processed foods including Italian-style salami, peanut butter, veggie booty and dry
cereal (7). Widespread distribution of these foods makes tracking the transmission of Salmonella
difficult during outbreak investigations. Outbreak investigations can also be hindered if the food
source is not listed on the investigation questionnaire (27). For example, for the 2008 Salmonella
outbreak due to consumption of Jalapeño and Serrano peppers, initial questionnaires did not ask if
peppers were recently eaten (6). In order to define the routes of transmission of Salmonella
within the food system, molecular subtyping methods have been employed to distinguish
outbreak clones from non-related clones (18).
Serotyping is one of the most common molecular subtyping methods for Salmonella.
Serotyping distinguishes Salmonella based on immunological classification of the H and O
antigens (21) and is typically the first subtyping method utilized during an outbreak. However,
serotyping alone cannot distinguish outbreak clones of Salmonella.
Several nucleic acids-based molecular subtyping methods have been used to subtype
Salmonella, including amplified fragment length polymorphism (AFLP) (20, 36, 39,45, 49),
49
multiple loci variable number tandem repeat analysis (MLVA) (2, 34, 35, 40), and pulsed-field
gel electrophoresis (PFGE) (13). PFGE is currently considered the ―gold standard‖ method for
subtyping foodborne pathogens and is the subtyping method used by PulseNet, the molecular
surveillance network in the U.S. and throughout the world to investigate foodborne illnesses and
outbreaks (19). The main advantage of PFGE is its high discriminatory power (i.e. ability to
separate unrelated strains) for subtyping foodborne pathogens, including many of the major
serovars of Salmonella (31). However, PFGE lacks discriminatory power for highly clonal
serovars of Salmonella, such as S. Enteritidis (19, 54) and S. Montevideo (9), or highly clonal
phage types like S. Typhimurium DT104 (19). The multistate S. Enteritidis outbreak associated
with shell eggs in 2010 was caused by the most common PFGE-XbaI pattern (JEGX01.0004) for
S. Enteritidis in the PulseNet database (8). A similar scenario was also observed recently during
the 2010 Italian-style salami outbreak, when the outbreak clone of S. Montevideo had the most
common PFGE pattern in the PulseNet database (9). Besides inadequate discriminatory power,
PFGE sometimes produces ambiguous data that are hard to compare and interpret between
different laboratories. To enhance comparability and interpretation, a standardized PFGE
protocol and an extensive quality assurance system were established by CDC (13, 19).
Compared to PFGE, multilocus sequence typing (MLST), which targets nucleotide
sequence differences of several DNA loci, has the potential to be a less labor-intensive method.
Moreover, DNA sequence data are discreet, unambiguous, highly informative, portable and
reproducible. Although MLST is an attractive subtyping approach, no satisfactory MLST scheme
has yet been developed for subtyping Salmonella during outbreak investigations. MLST schemes
targeting housekeeping genes have been developed; however, these schemes usually have much
lower discriminatory power than that of PFGE (16, 28, 33, 49). This suggests that housekeeping
genes do not provide sufficient resolution for investigating Salmonella outbreaks.
In order to increase discriminatory power, virulence genes have been included in MLST
schemes for subtyping Salmonella (17). Unlike housekeeping genes, virulence genes are
50
commonly under positive, diversifying selection (15). As a result, DNA sequences of virulence
genes tend to be more variable than housekeeping genes, and thus able to provide increased
discrimination (11, 17). Virulence genes have also been shown to provide high epidemiologic
concordance (i.e. able to group related strains together). For example, six virulence genes were
targeted in an MVLST (multi-virulence-locus sequence typing) scheme for subtyping Listeria
monocytogenes, which showed very high discriminatory power (0.99) and perfect epidemiologic
concordance (1.0) (11). Tankouo-Sandjong et al. (47) developed an MLST scheme based on both
virulence genes and housekeeping genes to identify serovars of Salmonella. This scheme targeted
the virulence genes fliC and fljB along with two housekeeping genes, gyrB and atpD (47).
However, the use of this MLST scheme to further characterize strains below serovar level was not
tested. In another MLST study, virulence genes and housekeeping genes showed high
discriminatory power (0.98) for subtyping S. Typhimurium, which was slightly higher than that of
PFGE (0.96) (17). In that MLST scheme three virulence genes, hilA, pefB and fimH, were
included together with the 16S rRNA gene and three housekeeping genes. Although this MLST
scheme appeared to have adequate discriminatory power for subtyping S. Typhimurium, its
capacity to discriminate strains from more clonal serovars such as S. Enteritidis was not tested.
Comparative genomic analysis (25) suggested that virulence genes alone are not discriminatory
enough for differentiating outbreak clones of S. Enteritidis. Therefore, additional genome targets
with greater sequence diversity than virulence genes are needed in order to create an effective
MLST scheme for Salmonella.
One of the fastest evolving genetic elements in bacteria genomes is CRISPRs (Clustered
Regularly Interspaced Palindromic Repeats) (43). CRISPRs have been identified within the
genomes of many archaeal and bacterial species, including Salmonella (30, 43, 50). CRISPRs
encode tandem sequences containing 21-47 bp direct repeats (DRs) separated by spacers of
similar size (Fig. 3.1). Spacers are derived from foreign nucleic acids such as phage or plasmids
and can protect bacteria from subsequent infection by homologous phage and plasmids (1).
51
Many CRISPR loci are flanked at the 3’ end by an AT-rich leader sequence and CRISPR-
associated (Cas) genes (Fig. 3.1) (1, 3, 26). As a bacterial immune system against foreign DNA,
CRISPRs must evolve rapidly to adapt to different phage pools (52). Besides addition of new
spacers, deletion of spacers is also frequently observed (12, 38). Because of the high
polymorphism of CRISPRs, they have been successfully used to subtype M. tuberculosis during
outbreak investigations (24). CRISPR sequence analysis has also characterized a number of other
bacteria, including Yersinia pestis (38), serotype M1 group A Streptococcus strains (29), and
Campylobacter jejuni (42).
Two CRISPR loci are found in all Salmonella serovars in the CRISPR database
(http://crispr.u-psud.fr/crispr/) (22, 50). Generally, the two CRISPR loci have different number
of repeats/spacers and different set of spacers. There have been no reports of CRISPRs being
used as markers in an MLST scheme for subtyping Salmonella. Therefore, the purpose of the
present study was to investigate whether MLST based on both virulence genes and CRISPRs can
accurately differentiate outbreak clones of the major serovars of Salmonella.
52
3.3 Materials and methods
Bacterial isolates and DNA extraction. All 171 Salmonella isolates used in this study
(Table 3.2) were from culture collections maintained by the Centers for Disease Control and
Prevention (CDC) in Atlanta, GA, USA. This set of isolates represents the 9 serovars most
commonly associated with human disease and includes isolates involved in multiple outbreaks,
with 2 to 3 isolates per outbreak. In some cases, isolates obtained from the same outbreak which
had different PFGE patterns (had poor epidemiologic concordance by PFGE) were deliberately
included. All isolates were previously analyzed by serotyping and most isolates were analyzed
by PFGE by CDC. Bacterial isolates were stored at -80°C in 20% glycerol. When needed,
isolates were grown overnight in Tryptic Soy Broth (TSB) (Difco Laboratories, Becton Dickinson,
Sparks, MD) at 37°C. For all isolates, DNA was extracted using the UltraClean Microbial DNA
extraction kit (Mo Bio Laboratories, Solana Beach, CA) and stored at -20°C before use.
Selection of virulence genes. Two virulence genes (fimH and sseL) and two Clustered
Regularly Interspaced Short Palindromic Repeats regions (CRISPR1 and CRISPR2) were
selected as markers for MLST. The lengths and functions of these MLST markers are listed in
Table 3.3. Additionally, 12 other virulence genes (hilA, fimH2, pipB, sopE, sseF, sseJ, siiA, sifB,
stdA, fimA, bcfC and phoQ) (Table S1) were initially investigated, but were excluded from the
MLST scheme due to inadequate sequence variation.
PCR amplification. Primers were designed using Primer 3.0
(http://frodo.wi.mit.edu/primer3/) and are listed in Tables 3.4 and S1. Primers for CRISPR1 were
designed based upon consensus alignments of the published S. Typhimurium LT2 (accession
number AE006468) and S. Newport str. SL254 genomes (accession number CP001113), and the
S. Javiana str. GA_MM04042433 (accession number ABEH00000000) whole genome shotgun
sequence (Table 3.4). Primers for the other three markers were designed based on the published S.
53
Typhimurium LT2 genome. PCR amplifications were performed using a Taq PCR Master Mix
Kit (Qiagen Inc., Balencia, CA) and a Mastercycler PCR thermocycler (Eppendorf Scientific,
Hamburg, Germany). A 25 µl PCR reaction system contained 12.5 µl Taq PCR 2×master mix,
9.5 µl PCR-grade water, 1.0 µl DNA template, 1.0 µl forward primer (final concentration, 0.4 µM)
and 1.0 µl reverse primer (final concentration, 0.4 µM). A single PCR cycling condition was
used for separately amplifying all four markers (initial denaturation at 94 °C for 10 min; 28
cycles of 94°C for 1 min, 55°C for 1 min,72°C for 1 min; final extension at 72°C for 10 min).
DNA sequencing. After PCR, products for sequencing were treated with 1/20 volume of
shrimp alkaline phosphatase (1 U/µl, USB Corp. Cleveland, OH) and 1/20 volume of exonuclease
I (10 U/µl, USB Corp). The mixture was then incubated at 37°C for 45 min to degrade remaining
primers and unincorporated dNTPs. After that, the mixture was incubated at 80°C for 15 min to
inactivate the added enzymes. PCR products were sent to the Genomics Core Facility at the
Pennsylvania State University for sequencing using the ABI Data 3730XL DNA Analyzer. In
order to obtain complete DNA sequences of fimH and sseL, two more primers targeting the
internal regions of these two genes were used together with the forward and reverse primers
(Table 3.4). Both DNA strands of the amplicons were sequenced.
Sequence analysis and sequence type assignment. For fimH and sseL, sequences were
aligned and single nucleotide polymorphisms (SNPs) were identified using MEGA 4.0 (46). For
CRISPR1 and CRISPR2, analyses of the spacer arrangements were performed using
CRISPRcompar (23) and spacers were visualized as described by Deveau et al. (12). Different
allelic types (ATs) (sequences with at least one-nucleotide difference or one-spacer difference in
the case of CRISPRs) were assigned arbitrary numbers. The combination of 4 alleles (fimH, sseL,
CRISPR1 and CRISPR2) determined its allelic profile and each unique allelic profile was
designated as a unique sequence type (ST).
54
Calculation of epidemiologic concordance (E). Epidemiologic concordance (E) was
calculated using the equation developed by the European Study Group on Epidemiologic Markers
(44).
Cluster analysis. Cluster analyses were performed based on allelic profile data and
results were visualized using the tree drawing tool on PubMLST (www.pubmlst.org). CRISPR1
and CRISPR2 were combined into one allele for a more accurate cluster analysis, because
CRISPR1 and CRISPR2 might be spatially linked (50).
Nucleotide sequence accession number. DNA sequences of the four genetic MLST
markers were deposited in GenBank under accession numbers HQ329797 to HQ329931.
55
3.4 Results
Results of MVLST. We began this study by sequencing 14 virulence genes (fimH, sseL,
hilA, fimH2, pipB, sopE, sseF, sseJ, siiA, sifB, stdA, fimA, bcfC and phoQ) from 20 S.
Typhimurium, 15 S. Newport, and 15 S. Enteritidis isolates. Two virulence genes, fimH and sseL,
were found to provide discrimination equal to the combined discrimination of all 14 virulence
genes (data not shown), therefore, the other 12 virulence genes were excluded from the rest of the
study. fimH and sseL were sequenced from the remaining isolates, and the total number of allelic
types was about the same for fimH (17 allelic types) and sseL (16 allelic types) (Table 3.5). The
total number of polymorphic sites and percentage of polymorphic sites for fimH was 48 and
4.76% and for sseL it was 69 and 7.23%, respectively (Table 3.6). For sequence variations of
fimH within each serovar, the percentage of polymorphic sites ranged from 0% to 1.79%. For
sseL, the percentage of polymorphic sites ranged from 0% to 3.88%. For both fimH and sseL,
less polymorphism was observed for serovars Typhimurium, Enteritidis, Heidelberg, Javiana and
I 4, [5], 12: i :-, compared to serovars Newport, Montevideo, Muenchen and Saintpaul (Table 3.6).
Sequences of sseL were especially conserved in serovars Typhimurium, Heidelberg, Javiana and I
4, [5], 12: i :-, with no SNPs observed within each serovar. For all serovars, a total of 39
polymorphic sites in sseL were nonsynonymous, and 13 polymorphic sites in fimH were
nonsynonymous (Table 3.6).
Addition of CRISPR1 and CRISPR2 in the MLST scheme. Since the discrimination
provided by virulence genes was limited (separation to outbreak level was not achieved), addition
of CRISPR1 and CRISPR2 into the MLST scheme was investigated. The number of allelic types
for CRISPR1 (49 allelic types) and CRISPR2 (53 allelic types) were significantly greater than
those for virulence genes (Table 3.5). In total, there were 69 sequence types based on both
virulence genes and CRISPRs for all 171 isolates (Table 3.5). An equal number of allelic types
56
was observed in both CRISPR1 and CRISPR2 for serovars Javiana and Montevideo (Table 3.5).
However, for serovars Typhimurium, Enteritidis, Newport, Heidelberg and Saintpaul, CRISPR2
contained more allelic types than CRISPR1. In contrast, for serovar Muenchen, CRISPR1
contained more allelic types than CRISPR2 (Table 3.5).
Repeat sequences of the two CRISPRs were generally conserved as shown by the typical
repeat in Table 3.7. However, SNPs were sometimes observed in the repeat sequences in both
CRISPRs and we define these as ―repeat variants‖ (Table 3.7). The repeat variant of CRISPR1
had one SNP at the first nucleotide, which is A instead of C (Table 3.7). Terminal repeat
sequences which are located furthest from the leader sequence (Fig. 3.1) had more SNPs than the
repeat variants’ sequences when compared to the typical repeat sequence (Table 3.7).
The total numbers of unique spacers in CRISPR1 and CRISPR2 for all 171 isolates
analyzed were 166 and 182, respectively (Table 3.8). The number of spacers in CRISPR1 ranged
from 3 spacers to 24, while the number of spacers in CRISPR2 ranged from 2 to 25 (Table 3.8
and Fig. S1). CRISPR2 had more spacers than CRISPR1 for all serovars except S. Muenchen, in
which CRISPR1 had more spacers than CRISPR2 (Table 3.8 and Fig. S1). The number of
spacers also varied between different serovars. For example, the average number of spacers in
CRISPR2 of S. Muenchen was 2.5, while the average number of spacers in CRISPR2 of S.
Typhimurium was 19.6 (Table 3.8).
Cluster analyses. Cluster diagrams based on allelic profiles were constructed using only
the two virulence genes (Fig. 3.2a) and also using virulence genes combined with CRISPRs (Fig.
3.2b), respectively. Again, significantly greater separation was provided by the addition of
CRISPR1 and CRISPR2 for all serotypes, compared to the separation provided by virulence
genes alone. MLST results showed high congruence with serotypes of Salmonella. On both
cluster trees, different serovars occupied distinct branches, except serovars Typhimurium and I 4,
[5], 12: i :- , which were clustered together. Also, three singletons Mvo ST3, Mcn ST12 and S
ST4 were observed on both cluster trees (Fig. 3.2a and ab).
57
Comparison of MLST with PFGE. Compared to PFGE, the addition of CRISPRs into
the present MLST scheme provided greater discrimination of outbreak clones of S. Enteritidis
(Table 3.2 and 3.5). Most isolates of S. Enteritidis (25 out of 34) had either XbaI and BlnI PFGE
profile (JEGX01.0005, JEGA26.0004) or (JEGX01.0004, JEGA26.0002) (Table 3.2). Isolates
SE1, SE2, SE23, SE18, SE17, SE20, SE32 and SE33 had the same PFGE profile (JEGX01.0005,
JEGA26.0004), but had two MLST sequence types (E ST1 and E ST 9) (Table 3.2). Also, the
PFGE profile (JEGX01.0004, JEGA26.0002), which included isolates SE6, SE7, SE8, SE9, SE15,
SE16, SE19, SE30, SE12, SE13, SE14, SE26, SE31, SE28, SE29, SE24 and SE34, were further
separated into five sequence types (E ST3, E ST4, E ST6, E ST7, E ST8) by MLST (Table 3.2).
However, in the case of some serovars (S. Newport, S. Typhimurium, S. I 4, [5], 12: i :-, S.
Montevideo) PFGE provided greater separation than MLST for strains associated with different
outbreaks. For example, PFGE separated S. I 4, [5], 12: i :- isolates (ST1, ST2 and ST3) of the
turkey potpie outbreak (cluster 0706PAJPX-1c) from isolates (ST14 and ST15) of cluster
0607INjpx-1c, while these isolates could not be distinguished by MLST (Table 3.2). This was
also the case for S. Typhimurium isolates from the Noble Farm raw milk outbreak and outbreak
cluster 0309ORJPX-1c (Table 3.2). Another example when PFGE was more discriminatory than
MLST was the raw chicken outbreak (cluster 0807AZJIX-1c) and the salami/pepper outbreak
(cluster 0908ORJIX-1) of S. Montevideo (Table 3.2).
For S. Heidelberg, the most accurate outbreak identification was achieved by combining
MLST and PFGE. MLST provided separation for the cruise ship outbreak (cluster 0607NYJF6-
1c) and religious camp outbreak (cluster 0607PAJF6-1c), and for the hummus outbreak (cluster
JF6X01.0032) and outbreak cluster 0702TNJF6-1c, which could not be distinguished by PFGE
(Table 3.2). However, PFGE separated the cruise ship outbreak from the hummus outbreak,
which were indistinguishable by MLST (Table 3.2).
For S. Saintpaul, both methods allowed accurate separation and identification of all
outbreaks due to this serovar (Table 3.2).
58
Comparison of MLST results with epidemiologic data. Isolates with the same cluster
code had identical MLST sequence types for serovars S. Typhimurium, S. Newport, S. I 4, [5], 12:
i :- , S. Saintpaul and S. Montevideo (Table 3.2). MLST sequence types were the same among
isolates with the same cluster code for S. Enteritidis, except clusters 0505GAJEC-1c and
0612MEJEC-1c, and S. Heidelberg, except for 0704AZJPX-1c. Isolates SE10 and SE11, which
have the same cluster code (0505GAJEC-1c), had different sequence types and also different
PFGE patterns (Table 3.2). Three isolates of S. Enteritidis (SE12, SE13 and SE14) and three
isolates of S. Heidelberg (SH18, SH19 and SH20) which have the same cluster code
(0612MEJEC-1c and 0704AZJPX-1c), respectively, also had different sequence types (Table 3.2).
For S. Muenchen, almost all isolates within each of the six cluster codes had different sequence
types (Table 3.2). We could not perform a similar analysis with S. Javiana, because it did not
contain any isolates with a cluster code identified by PulseNet.
Epidemiologic concordance of MLST. Values of epidemiologic concordance of MLST
and PFGE for each serovar are listed in Table 3.9, except for the serovar Javiana which didn’t
contain any isolates with a cluster code identified by PulseNet which prevented calculation of an
epidemiologic concordance value for this serovar. Epidemiologic concordance values were
calculated based on isolates from well-defined outbreaks (isolates with cluster codes), so sporadic
isolates and isolates without cluster codes were excluded from epidemiologic concordance
calculations. Values of epidemiologic concordance were biased against PFGE, because isolates
from outbreaks with variations in PFGE patterns were deliberately included in this study, which
reduced the epidemiologic concordance of PFGE. For instance, isolates ST6, ST7 and ST8 were
all from the 2008 peanut butter outbreak, but each of them had a distinct PFGE pattern (Table
3.2). Generally speaking, MLST showed high epidemiologic concordance for subtyping all
serovars included in this study, except for S. Muenchen (epidemiologic concordance= 0.39)
(Table 3.9). MLST showed higher epidemiologic concordance than PFGE for serovars
Enteritidis, Typhimurium, Newport and Montevideo, equal epidemiologic concordance for
59
serovar Saintpaul, but lower epidemiologic concordance for serovars Heidelberg and Muenchen
(Table 3.9).
60
Table 3.1. Top nine most frequently reported serovars from human sources in 2005 which were
analyzed in the present study
Rank Serovar No. of laboratory-confirmed cases1
% of total cases
1 Typhimurium 6982 19.3
2 Enteritidis 6730 18.6
3 Newport 3295 9.1
4 Heidelberg 1903 5.3
5 Javiana 1324 3.7
6 I 4, [5], 12: i :- 822 2.3
7 Montevideo 809 2.2
8 Muenchen 733 2
9 Saintpaul 683 1.9
total 64.4
1Laboratory-confirmed cases include both outbreak cases and sporadic cases.
Data reproduced from CDC’s Salmonella annual review
(http://www.cdc.gov/ncidod/dbmd/phlisdata/salmtab/2006/SalmonellaAnnualSummary2006.pdf).
61
Table 3.2. Outbreak information, PFGE profile and MLST results for the 171 isolates analyzed in
the present study
CDC
Code1 Source State Food vehicle Cluster PFGE XbaI PFGE BlnI MLST
ST2
ST29 Water filter UT Frog 0909MAJPX-1 JPXX01.0177 JPXA26.0459 T ST1
ST30 Human /Stool MD Frog 0909MAJPX-1 JPXX01.0177 JPXA26.0459 T ST1
ST31 Human /Stool OH Frog 0909MAJPX-1 JPXX01.0177 JPXA26.0459 T ST1
ST4 Human /Stool CO Water 0803COJPX-1c JPXX01.0002 JPXA26.0002 T ST2
ST5 Water CO Water 0803COJPX-1c JPXX01.0002 JPXA26.0002 T ST2
ST6 Human /Stool OH peanut butter 0811MLJPX-1c JPXX01.0459 JPXA26.0462 T ST3
ST7 Human /Stool OH peanut butter 0811MLJPX-1c JPXX01.1825 JPXA26.0462 T ST3
ST8 Food/peanut butter MN peanut butter 0811SDCJPX-1c JPXX01.1818 JPXA26.0462 T ST3
ST9 Stool MA Raw milk Noble Farm outbreak JPXX01.0083 JPXA26.0019 T ST4
ST10 Raw milk MA Raw milk Noble Farm outbreak JPXX01.0083 JPXA26.0019 T ST4
ST17 NA OR NA 0309ORJPX-1c JPXX01.0981 JPXA26.0174 T ST4
ST18 NA OR NA 0309ORJPX-1c JPXX01.0981 JPXA26.0174 T ST4
ST11 Stool NM NA Santa Fe JPXX01.0003 JPXA26.0007 T ST5
ST12 Stool NM NA Santa Fe JPXX01.0003 JPXA26.0007 T ST5
ST13 Stool NM NA Santa Fe JPXX01.0003 JPXA26.0008 T ST5
ST26 Human /Stool OR Snake / mouse 0908ORJPX-1 JPXX01.0003 JPXA26.0003 T ST5
ST27 Human /Stool OR Snake / mouse 0908ORJPX-1 JPXX01.0003 JPXA26.0003 T ST5
ST28 Animal OR Snake / mouse 0908ORJPX-1 JPXX01.0003 JPXA26.0003 T ST5
ST39 Human /Stool VA Sporadic Sporadic JPXX01.0003 JPXA26.0042 T ST5
ST16 Stool MA Veggie booty 0704WIWWS-c JPXX01.1037 JPXA26.0333 T ST6
ST19 Stool VT Veggie booty 0704WIWWS-1c JPXX01.1037 JPXA26.0333 T ST6
ST20 Stool VT Veggie booty 0704WIWWS-1c JPXX01.1037 JPXA26.0333 T ST6
ST32 Human /Stool AR Day care 0602ARJPX-2c JPXX01.0010 JPXA26.0233 T ST7
ST33 Human /Stool AR Day care 0602ARJPX-2c JPXX01.0010 JPXA26.0233 T ST7
ST34 Human /Stool AR Day care 0602ARJPX-2c JPXX01.0010 JPXA26.0233 T ST7
ST40 Human /Stool NY Sporadic Sporadic JPXX01.0003 JPXA26.0042 T ST8
SE1 Human/stool MN Stuffed chicken 0603MNJEG-1c JEGX01.0005 JEGA26.0004 E ST1
SE2 Human/stool MN Stuffed chicken 0603MNJEG-1c JEGX01.0005 JEGA26.0004 E ST1
SE23 Human/stool MN NA 0603MNJEG-1c JEGX01.0005 JEGA26.0004 E ST1
SE18 Human/Stool MN NA 0803MNJEG-1 JEGX01.0005 JEGA26.0004 E ST1
SE3 Environment CA Almonds Almonds 2001 JEGX01.0012 NA E ST2
SE4 Food/raw almonds CA Almonds Almonds 2001 JEGX01.0012 NA E ST2
SE5 Environment CA Almonds Almonds 2001 JEGX01.0012 NA E ST2
SE21 Environment NA NA Almonds 2001 JEGX01.0013 NA E ST2
SE25 Environment NA Prison Almonds 2001 JEGX01.0013 NA E ST2
SE6 Human/stool ME NA 0612MEJEG-1c JEGX01.0004 JEGA26.0002 E ST3
SE7 Human/stool ME NA 0612MEJEG-1c JEGX01.0004 JEGA26.0002 E ST3
SE26 Human/stool CO NA NA JEGX01.0004 JEGA26.0002 E ST3
SE31 Human/stool CO NA NA JEGX01.0004 JEGA26.0002 E ST3
SE24 Human/Stool WV NA NA JEGX01.0004 JEGA26.0002 E ST3
SE8 Human/Stool PA Egg 0801PAJEG-1 JEGX01.0004 JEGA26.0002 E ST4
SE9 Human/Stool PA Egg 0801PAJEG-1 JEGX01.0004 JEGA26.0002 E ST4
SE15 Human/Stool PA NA 0801PAJEG-1 JEGX01.0004 JEGA26.0002 E ST4
SE34 Human/Stool CT NA NA JEGX01.0004 JEGA26.0002 E ST4
SE11 Human/stool GA Hospital eggs 0505GAJEG-1c JEGX01.0018 JEGA26.0005 E ST4
SE10 Human/stool GA Hospital eggs 0505GAJEG-1c JEGX01.0034 JEGA26.0005 E ST5
SE12 NA ME NA 0612MEJEG-1c JEGX01.0004 JEGA26.0002 E ST6
SE13 Human/stool ME NA 0612MEJEG-1c JEGX01.0004 JEGA26.0002 E ST6
SE14 Human/stool ME NA 0612MEJEG-1c JEGX01.0004 JEGA26.0002 E ST7
SE16 Human/stool GA NA 0506GAJEG-1c JEGX01.0004 JEGA26.0002 E ST8
SE19 Human/stool GA NA 0506GAJEG-1c JEGX01.0004 JEGA26.0002 E ST8
SE30 Human/stool GA Prison 0506GAJEG-1c JEGX01.0004 JEGA26.0002 E ST8
SE22 Human/stool OR NA 0509ORJEG-1c JEGX01.0004 JEGA26.0025 E ST8
SE27 Human/stool OR NA 0509ORJEG-1c JEGX01.0004 JEGA26.0025 E ST8
SE28 Human SC NA 0504SCJEG-1c JEGX01.0004 JEGA26.0002 E ST8
SE29 Human/stool ID NA 0504CAOCJEG-1c JEGX01.0004 JEGA26.0002 E ST8
62
SE17 NA OH Frozen chicken Outbreak 2005-28-076 JEGX01.0005 JEGA26.0004 E ST9
SE20 NA OH NA Outbreak 2005-28-076 JEGX01.0005 JEGA26.0004 E ST9
SE32 Human/Stool MI NA 0708MIJEG-1c JEGX01.0005 JEGA26.0004 E ST9
SE33 Human/Stool MI NA 0708MIJEG-1c JEGX01.0005 JEGA26.0004 E ST9
SN1 NA IL NA NA JJPX01.0014 NA N ST1
SN2 NA IL NA NA JJPX01.0014 NA N ST1
SN3 NA NA NA 0509NHJJP-1c. JJPX01.0061 JJPA26.0021 N ST2
SN4 NA NA NA 0509NHJJP-1c. JJPX01.0061 JJPA26.0021 N ST2
SN5 NA NA NA NA JJPX01.0001 NA N ST3
SN6 NA NA NA NA JJPX01.0001 NA N ST3
SN7 Human/Stool CA NA 0710CAJJP-1c JJPX01.0422 JJPA26.0196 N ST4
SN8 Human/Stool CA NA 0710CAJJP-1c JJPX01.0422 JJPA26.0196 N ST4
SN11 Human/Stool SD NA 0712SDJJP-1c JJPX01.0654 JJPA26.0208 N ST4
SN12 Human/Stool SD NA 0712SDJJP-1c JJPX01.0654 JJPA26.0208 N ST4
SN9 Human/Stool AZ NA 0802AZJJP-1c JJPX01.0696 JJPA26.0212 N ST5
SN10 Human/Stool AZ NA 0802AZJJP-1c JJPX01.0438 JJPA26.0212 N ST5
SN13 Human/Stool GA NA 0711GAJJP-1c JJPX01.1319 JJPA26.0542 N ST6
SN14 Human/Stool GA NA 0711GAJJP-1c JJPX01.1319 JJPA26.0542 N ST6
SN15 Human/Stool GA NA 0711GAJJP-1c JJPX01.1319 JJPA26.0542 N ST6
SH1 Human DE cruise ship 0607NYJF6-1c JF6X01.0022 NA H ST1
SH2 Human NY cruise ship 0607NYJF6-1c JF6X01.0022 NA H ST1
SH3 Human NY cruise ship 0607NYJF6-1c JF6X01.0022 NA H ST1
SH8 Human IL hummus 0707ILJF6-1c JF6X01.0032 JF6A26.0076 H ST1
SH9 Human IL hummus 0707ILJF6-1c JF6X01.0032 JF6A26.0076 H ST1
SH10 Human IL hummus 0707ILJF6-1c JF6X01.0032 JF6A26.0076 H ST1
SH11 Human IL hummus 0707ILJF6-1c JF6X01.0032 JF6A26.0076 H ST1
SH16 NA NA Sporadic Sporadic JF6X01.0122 NA H ST1
SH17 NA NA Sporadic Sporadic JF6X01.0022 NA H ST1
SH18 Human NA NA 0704AZJPX-1c JF6X01.0022 NA H ST1
SH4 Human PA a religious camp 0607PAJF6-1c JF6X01.0022 NA H ST2
SH5 Human PA a religious camp 0607PAJF6-1c JF6X01.0022 NA H ST2
SH6 Human PA a religious camp 0607PAJF6-1c JF6X01.0022 NA H ST2
SH7 Human PA a religious camp 0607PAJF6-1c JF6X01.0022 NA H ST2
SH15 NA NA Sporadic Sporadic JF6X01.0051 NA H ST2
SH12 Human TN NA 0702TNJF6-1c JF6X01.0032 JF6A26.0076 H ST3
SH13 Human TN NA 0702TNJF6-1c JF6X01.0032 JF6A26.0076 H ST3
SH14 NA NA Sporadic Sporadic JF6X01.0135 NA H ST4
SH19 Human NA NA 0704AZJPX-1c JF6X01.0022 NA H ST5
SH20 Human NA NA 0704AZJPX-1c JF6X01.0022 NA H ST6
SJ1 NA AL NA NA JGGX01.0012 NA J ST1
SJ5 NA AR NA NA JGGX01.0012 NA J ST1
SJ13 NA LA NA NA NA NA J ST1
SJ15 NA
outbreak NA JGGX01.0036 JGGA26.0017 J ST1
SJ2 NA TX NA NA JGGX01.0213 NA J ST2
SJ3 NA LA NA NA NA NA J ST3
SJ8 NA LA NA NA NA NA J ST3
SJ4 NA TX NA NA NA NA J ST4
SJ6 NA AR NA NA JGGX01.0179 NA J ST5
SJ9 NA AR NA NA JGGX01.1226 NA J ST5
SJ7 NA TX NA NA JGGX01.1525 NA J ST6
SJ10 NA HU NA NA NA NA J ST7
SJ11 NA MD NA NA JGGX01.0362 NA J ST8
SJ12 NA IL NA NA JGGX01.1352 NA J ST9
SJ14 NA NV NA NA NA NA J ST10
ST15 Stool CA Turkey potpie 0706PAJPX-1c JPXX01.0206 JPXA26.0180 I ST1
ST25 Stool GA Turkey potpie 0706PAJPX-1c JPXX01.0206 JPXA26.0180 I ST1
ST35 Food/Turkey potpie WI Turkey potpie 0706PAJPX-1c JPXX01.0206 JPXA26.0180 I ST1
ST145 Stool IN NA4 0607INjpx-1c JPXX01.0621 JPXA26.0160 I ST1
ST155 Stool IN NA 0607INjpx-1c JPXX01.0621 JPXA26.0160 I ST1
ST215 Human /Stool OH Snake 0806OHJPX-1c JPXX01.1596 JPXA26.0491 I ST2
ST225 Human /Stool OH Snake 0806OHJPX-1c JPXX01.1596 JPXA26.0491 I ST2
ST235 Human /Stool OH Snake 0806OHJPX-1c JPXX01.1596 JPXA26.0491 I ST2
63
ST245 Food/Egg wash ME Egg 0404PAJPX-1c JPXX01.0621 JPXA26.0057 I ST3
ST255 NA VT Egg 0404PAJPX-1c JPXX01.0621 JPXA26.0057 I ST3
ST355 Human /Stool OH Sporadic Sporadic JPXX01.0621 JPXA26.0055 I ST4
ST365 Human /Stool MA Sporadic Sporadic JPXX01.1212 JPXA26.0108 I ST4
ST375 Human /Stool MO Sporadic Sporadic JPXX01.0206 JPXA26.0380 I ST4
SMvo1 Blood TX NA NA NA NA Mvo ST1
SMvo2 Stool TX NA NA NA NA Mvo ST2
SMvo7 Human MD NA NA JIXX01.0524 NA Mvo ST3
SMvo3 Human/Rectal swab AZ Raw chicken 0807AZJIX-1c JIXX01.1014 NA Mvo ST3
SMvo8 Human/Stool AZ Raw chicken 0807AZJIX-1c JIXX01.0126 JIXA26.0012 Mvo ST3
SMvo9 Human/Swab AZ Raw chicken 0807AZJIX-1c JIXX01.0126 JIXA26.0012 Mvo ST3
SMvo10 Human/Stool AZ Raw chicken 0807AZJIX-1c JIXX01.0126 JIXA26.0012 Mvo ST3
SMvo11 Human/Stool UT Salami/pepper 0908ORJIX-1 JIXX01.0011 JIXA26.0012 Mvo ST3
SMvo12 Human/Urine OR Salami/pepper 0908ORJIX-1 JIXX01.0011 JIXA26.0012 Mvo ST3
SMvo13 Human/Stool AZ Salami/pepper 0908ORJIX-1 JIXX01.0011 NA Mvo ST3
SMvo15 Human/Stool TN Salami/pepper 0908ORJIX-1 JIXX01.0011 NA Mvo ST3
SMvo14 NA AZ NA NA NA NA Mvo ST3
SMvo4 NA TX NA NA JIXX01.0388 NA Mvo ST4
SMvo5 Human/Stool TX NA NA JIXX01.0875 NA Mvo ST5
SMvo6 Human/Stool TN NA NA JIXX01.1005 NA Mvo ST6
SMcn1 NA TX NA outbreak JJPX01.0014 NA Mcn ST1
SMcn2 NA NYC NA outbreak JJPX01.0014 NA Mcn ST2
SMcn3 Human/Stool LA NA 0509NHJJP-1c. JJPX01.0061 JJPA26.0021 Mcn ST3
SMcn4 NA TX NA 0509NHJJP-1c. JJPX01.0061 JJPA26.0021 Mcn ST4
SMcn5 NA TX NA NA NA NA Mcn ST5
SMcn6 Human/Stool TX NA NA NA NA Mcn ST6
SMcn7 NA TX NA 0710CAJJP-1c JJPX01.0422 JJPA26.0196 Mcn ST7
SMcn8 NA TX NA 0710CAJJP-1c JJPX01.0422 JJPA26.0196 Mcn ST8
SMcn9 Human/Stool TX NA 0802AZJJP-1c JJPX01.0696 JJPA26.0212 Mcn ST9
SMcn10 NA TX NA 0802AZJJP-1c JJPX01.0438 JJPA26.0212 Mcn ST10
SMcn11 Human/Stool TX NA 0712SDJJP-1c JJPX01.0654 JJPA26.0208 Mcn ST11
SMcn12 Human MD NA 0712SDJJP-1c JJPX01.0654 JJPA26.0208 Mcn ST12
SMcn13 NA OR Orange Juice 0711GAJJP-1c JJPX01.1319 JJPA26.0542 Mcn ST13
SMcn15 NA WA Orange Juice 0711GAJJP-1c JJPX01.1319 JJPA26.0542 Mcn ST13
SMcn14 NA WA Orange Juice 0711GAJJP-1c JJPX01.1319 JJPA26.0542 Mcn ST14
SS10 Human MA NA 0806MAJN6-1c JN6X01.0034 JN6A26.0038 S ST1
SS11 Human MA NA 0806MAJN6-1c JN6X01.0034 JN6A26.0038 S ST1
SS12 Human MA NA 0806MAJN6-1c JN6X01.0034 JN6A26.0038 S ST1
SS6 Human NE sprouts 0902NEJN6-1 JN6X01.0072 NA S ST2
SS7 Human NE sprouts 0902NEJN6-1 JN6X01.0072 NA S ST2
SS8 Human NE sprouts 0902NEJN6-1 JN6X01.0072 NA S ST2
SS9 Human NE sprouts 0902NEJN6-1 JN6X01.0072 NA S ST2
SS1 NA MN jalapeños 0805NMJN6-1c JN6X01.0048 JN6A26.0019 S ST3
SS2 Human TX jalapeños 0805NMJN6-1c JN6X01.0048 JN6A26.0019 S ST3
SS3 Human NM jalapeños 0805NMJN6-1c JN6X01.0048 JN6A26.0019 S ST3
SS4 Human AZ jalapeños 0805NMJN6-1c JN6X01.0048 JN6A26.0019 S ST3
SS18 NA NE Sporadic Sporadic JN6X01.0622 NA S ST3
SS19 NA TX Sporadic Sporadic JN6X01.0067 JN6A26.0001 S ST3
SS16 NA CA Sporadic Sporadic JN6A26.0026 NA S ST4
SS13 NA CA NA 0807LACJN6-1c JN6X01.0021 JN6A26.0019 S ST5
SS14 NA CA NA 0807LACJN6-1c JN6X01.0021 JN6A26.0019 S ST5
SS15 NA MD Sporadic Sporadic JN6X01.0170 NA S ST5
SS20 NA NV Sporadic Sporadic JN6X01.0623 JN6A26.0047 S ST6
1ST: S. Typhimurium (ST 29-31 are isolates of S. Typhimurium var Copenhagen). SE: S.
Enteritidis. SN: S. Newport. SH: S. Heidelberg. SJ: S. Javiana. SI: S. I 4, [5], 12; i: -. SMvo: S.
Montevideo. SMcn: S. Muenchen. SS: S. Saintpaul.
64
2ST: sequence type. T: S. Typhimurium. E: S. Enteritidis. N: S. Newport. H: S. Heidelberg. J:
S. Javiana. Mvo: S. Montevideo. Mcn: S. Muenchen. S: S. Saintpaul. For instance, T ST1
stands for Typhimurium sequence type 1.
4NA: Not available.
5ST1-3, 14-15,21-25 and 35-37 are isolates of S. I 4, [5], 12; i: -.
65
Table 3.3. Size, function and nucleotide location of the four markers targeted in the present study
Marker Size (bps) Function Nucleotide location in S.
Typhimurium LT2
fimH 1008 Host-cell-specific recognition 28,425 - 29,432
sseL 954 Inflammation and macrophage killing 2,394,795 - 2,395,748
CRISPR1 122-8541
Defense against phage 3,076,611 - 3,077,006
CRISPR2 183-15251
Defense against phage 3,094,279 - 3,096,260
1 Length of CRISPRs varied because the number of repeats/spacers changed among the different
strains analyzed.
66
Table 3.4. Primers used to amplify and sequence the four MLST markers
Marker Primer sequence (5'-3') Note
fimH CGTCGTCATAAAAGGAAAAA Forward primer for both amplification and sequencing
GAACAAAACACAACCAATAGC Reverse primer for both amplification and sequencing
CTCGCCAGACAATGTTTACT Reverse primer for sequencing internal region
CATTCACTTCGCAGTTTTG Forward primer for sequencing internal region
sseL AGGAAACAGAGCAAAATGAA Forward primer for both amplification and sequencing
TAAATTCTTCGCAGAGCATC Reverse primer for both amplification and sequencing
GGAGTTGAAAATCTTTGGTG Reverse primer for sequencing internal region
TTTACCGAGAGAAAAGGTGA Forward primer for sequencing internal region
CRISPR1 GATGTAGTGCGGATAATGCT Forward primer for both amplification and sequencing
GGTTTCTTTTCTTCCTGTTG 1Reverse primer for both amplification and sequencing
GATGATATGGCAACAGGTTT 1Reverse primer for both amplification and sequencing
TATTGACTGCGATGAGATGA 2Reverse primer for both amplification and sequencing
CRISPR2 ACCAGCCATTACTGGTACAC Forward primer for both amplification and sequencing
ATTGTTGCGATTATGTTGGT Reverse primer for both amplification and sequencing
1 The 2 reverse primers (reverse 1 and reverse 2) of CRISPR1 were added together with forward
primer to amplify CRISPR1 in all serovars except S. Javiana.
2 Reverse primer for SJ (S. Javiana) was needed for amplification and sequencing of CRISPR1 in
S. Javiana isolates.
67
Table 3.5. Number of isolates, allelic types and sequence types in each serovar
Serovar No. of Isolates No. of allelic types No. of MLST
STs No. of PFGE patterns
fimH sseL CRISPR1 CRISPR2
Typhimurium 26 3 1 7 8 8 13
Enteritidis 34 2 3 2 6 9 7
Newport 15 3 4 4 6 6 8
Heidelberg 20 2 1 1 5 6 5
Javiana 15 3 1 10 10 10 8
I 4, [5], 12; i: - 13 1 1 1 4 4 7
Montevideo 15 2 2 6 6 6 7
Muenchen 15 2 2 14 2 14 7
Saintpaul 18 2 2 5 6 6 10
Total 171 171
16 49 53 69 72
1Total number of allelic types for fimH does not equal the sum of allelic types in each serovar,
because the same allelic type was sometimes present in more than one serovar.
68
Table 3.6. Allelic polymorphisms and nucleotide substitutions in the nucleotide sequences of
fimH and sseL
Gene Serovar No. of
polymorphic sites
% of
polymorphic sites
No. of
synonymous substitutions
No. of
nonsynonymous substitutions
fimH Typhimurium 2 0.2 1 1
Enteritidis 1 0.1 0 1
Newport 10 0.99 6 4
Heidelberg 1 0.1 1 0
Javiana 2 0.2 0 2
I 4, [5], 12; i: - 0 0 0 0
Montevideo 13 1.29 10 3
Muenchen 16 1.59 13 3
Saintpaul 18 1.79 14 4
Total 48 4.76 35 13
sseL Typhimurium 0 0 0 0
Enteritidis 2 0.21 1 1
Newport 18 1.89 8 10
Heidelberg 0 0 0 0
Javiana 0 0 0 0
I 4, [5], 12; i: - 0 0 0 0
Montevideo 10 1.05 4 6
Muenchen 6 0.63 3 3
Saintpaul 37 3.88 15 22
Total 69 7.23 30 39
69
Table 3.7. Analysis of CRISPR repeat sequences
CRISPR Type Repeat sequences (5’-3’)1
CRISPR1 Typical repeat CGGTTTATCCCCGCTGGCGCGGGGAACAC
Repeat variant AGGTTTATCCCCGCTGGCGCGGGGAACAC
Terminal repeats GTGTTTATCCCCGCTGACGCGGGGAACAC
GTGTTTATCCCCGCTGGCGCGGGGAACAT
CRISPR2 Typical repeat Same as the typical repeat in CRISPR1
Repeat variants GGGTTTATCCCCGCTGGCGCGGGGAACAC
CAGTTTATCCCCGCTGGCGCGGGGAACAC
CGGTTTATCCCCGCTGACGCGGGGAACAT
CGGTTTATCCCCGCTAGCGCGGGGAACAC
CGGTTTATCCCCGCTGACGCGGGGAACAC
TGGTTTATCCCCGCTGGCGCGGGGAACAC
CGGTTTATCCCCGCTGGCACGGGGAACAC
CGATTTATCCCTGCTGGCGCGGGGAACAC
CGGTTTATCCCTGCTGGCGCGGGGAACAC
Terminal repeats ACGGCTATCCTTGTTGGCGCGGGGAACAC
CGGTTTATCCCCGCTGCGCGGGGAACACT
1Underscored nucleotides are SNPs, compared to the typical repeat.
70
Table 3.8. Analysis of CRISPR spacers in different serovars
Serovar No. of unique spacers
Avg no. of
spacers + SD1
Minimum no. of spacers
Maximum no. of spacers
CRISPR1 CRISPR2 CRISPR1 CRISPR2 CRISPR1 CRISPR2 CRISPR1 CRISPR2
Typhimurium 26 34 11.4+4.0 19.6+6.8 3 4 14 25
Enteritidis 9 10 8.5+0.6 8.8+1.6 8 7 9 10
Newport 31 43 11.3+4.9 16.3+3.4 4 10 14 19
Heidelberg 10 18 10.0+0.0 12.6+2.7 10 10 10 17
Javiana 9 16 6.4+2.0 9.4+4.0 4 2 9 14
I 4, [5], 12; i: - 13 23 13+0 24+1 13 13 23 25
Montevideo 38 40 13.2+5.6 17.7+3.0 9 14 24 22
Muenchen 34 5 12.8+5.0 2.5+0.7 6 2 20 3
Saintpaul 35 33 12.2+1.3 16.5+5.6 11 7 14 23
Total2
166 182 10.8+4.5 14.4+6.4
1 SD: value of standard deviation.
2 Number of total unique spacers does not equal the sum of unique spacers in each serovar,
because a unique spacer was sometimes present in more than one serovar.
71
Table 3.9. Comparison of epidemiologic concordance1 between PFGE and MLST based on
virulence genes and CRISPRs for the selected strains analyzed in the present study
Subtyping
method
Enteritidis Typhimurium Newport Heidelberg I 4, [5], 12; i: -
Saintpaul Montevideo Muenchen
MLST 0.94 1.00 1.00 0.88 1.00 1.00 1.00 0.39
PFGE2 0.91 0.91 0.93 1.00 1.00 1.00 0.87 0.92
1Values for epidemiologic concordance were calculated based on isolates with cluster codes
identified by PulseNet.
2 The above values for epidemiologic concordance are biased against PFGE, because in some
cases outbreaks that contained strains with variations in PFGE patterns (had poor epidemiologic
concordance by PFGE) were deliberately selected in the present study.
72
Figure 3.1. Schematic view of the two CRISPR systems in Salmonella Typhimurium LT2.
Direct repeats and spacers are represented by black diamonds and white rectangles, respectively.
The terminal direct repeats are represented by white diamonds. L stands for leader sequence. cas
genes are in grey while other core flanking genes (ygcF, iap and ptps) are in white. The figure is
not drawn to scale.
CRISPR1
CRISPR2
5
’
5
’
3
’
’
73
Figure 3.2. (a) Cluster diagram based on only fimH and sseL. (b) Cluster diagram based on fimH,
sseL and CRISPRs (combined allele of CRISPR1 and CRISPR2).
ST: sequence type. T: Typhimurium. E: Enteritidis. N: Newport. H: Heidelberg. J: Javiana. I:
I 4, [5], 12: i :-. Mvo: Montevideo. Mcn: Muenchen. S: Saintpaul. CRISPR1 and CRISPR2
were combined into one allele for the cluster analysis because CRISPR1 and CRISPR2 are
spatially linked (50).
(b)
(a)
74
3.5 Discussion
There are several important criteria to follow when selecting genetic markers to use in an
MLST scheme. First, the selected genetic markers should exhibit adequate sequence variations to
provide separation of unrelated strains (37). Secondly, genetic markers which provide
epidemiologically meaningful information should be selected so that the MLST scheme can
exhibit high epidemiologic concordance. Last but not least, genetic markers should be present in
the genome within all strains of the species of interest. Previous studies demonstrated that MLST
schemes based on Salmonella housekeeping genes showed poor discriminatory power when
compared to PFGE (16, 28, 49). Inclusion of virulence genes into one published MLST scheme
for subtyping S. Typhimurium increased discriminatory power to 0.98, which was comparable to
that of PFGE (0.96) (17). Virulence genes provided epidemiologically meaningful separation and
clustering of strains of Listeria monocytogenes (11). Besides virulence genes, CRISPRs were
selected as markers in the current MLST scheme because they were found to be one of the fastest
evolving genetic elements in bacterial genomes (43).
In the present study, cluster analyses based on the two virulence genes and two CRISPRs
accurately grouped isolates according to their specific serovars, except serovar Typhimurium and
I 4, [5], 12: i :- , which were clustered together. As serovar I 4, [5], 12: i:- is a monophasic
variant of serovar S. Typhimurium (14), our result is not unexpected. Virulence genes were
previously found to provide accurate identification of different serovars of Salmonella in other
studies as well (41, 47, 48).
Addition of CRISPRs significantly increased discriminatory power (Fig. 3.2) compared
to previously published MLST schemes, and the identification of individual outbreak clones was
achieved (Table 3.2). For example, one MLST scheme based on three housekeeping genes
75
(manB, pduF, and glnA) genes and one virulence gene (spaM) identified one sequence type
among 85 S. Typhimurium isolates and discriminatory power for the MLST scheme was 0 (16).
Another MLST scheme targeted seven housekeeping genes, aroC, dnaN, hemD, hisD, purE, sucA,
and thrA, and identified 12 sequence types among a total of 81 S. Newport isolates, which also
resulted in poor discriminator y power (0.61) (28). One MLST study based on virulence genes
(hilA, pefB and fimH), 16S rRNA gene and housekeeping genes showed high discriminatory
power (0.98) for subtyping S. Typhimurium (17); however, its capacity to discriminate strains
from more clonal serovars such as S. Enteritidis was not tested. In conclusion, the MLST scheme
described in the present study has superior discriminatory power, compared to previously
published MLST schemes for subtyping the major serovars of Salmonella, especially for the
highly clonal serovar S. Enteritidis.
As mentioned previously, the isolates selected for this study were biased towards those
that had poor epidemiologic concordance of PFGE, so future studies comparing of MLST and
PFGE need to be performed using a nonbiased strain collection. Generally speaking though, the
current MLST scheme showed high epidemiologic concordance for subtyping the major serovars
of Salmonella, except Muenchen (E=0.39) (Table 3.9). All S. Muenchen isolates had different
sequence types, except the two isolates, SMcn13 and SMcn15 from the orange juice outbreak
(Table 3.2). Interestingly, the allelic types of fimH and sseL were the same for all the S.
Muenchen isolates except isolate SMcn12 (Fig. 3.2a), which means CRISPR1 and CRISPR2
provided almost all of the discriminatory power in the case of S. Muenchen isolates (Fig. 3.2b).
Perhaps PFGE lacks pattern diversity for S. Muenchen because it cannot detect the subtle, but
epidemiologically important changes, detected by CRISPRs. Alternatively, CRISPRs may be
evolving too fast for S. Muenchen outbreak investigations, either because the specific niche
where S. Muenchen resides harbors a large number of different phage, and/or phage pools of S.
Muenchen are very dynamic. Dramatic differences have been observed in the rate of spacer
76
acquisition between different eubacteria. In Streptococcus thermophilus, CRISPRs are very
active and new spacer acquisition appear to be the primary mechanism of this species to defend
against phage (12); however, the rate of new spacer acquisition in other bacteria such as E. coli
appear to be much slower (50).
The current MLST scheme provided greater separation of S. Enteritidis isolates than
PFGE (Table 3.2). The predominant PFGE XbaI patterns for S. Enteritidis in the PulseNet
database are JEGX01.0004 and JEGX01.0005, which is problematic because this lack of PFGE
pattern diversity sometimes makes it difficult to separate potential outbreak-related isolates from
sporadic isolates (19). The discriminatory power of PFGE has been increased by the combination
of multiple restriction enzymes (54). However, whether the increased discrimination caused
potential loss of epidemiologic concordance was not addressed in that study. The present MLST
scheme allowed separation of the two predominant PFGE patterns of S. Enteritidis isolates (Table
3.2) and resulted in high epidemiologic concordance (Table 3.9). CRISPRs provided most of the
discrimination (Fig. 3.2b). CRISPRs in S. Enteritidis are evolving due to plasmids and/or phage
present in the environment (52). Fortunately, the rate of spacer insertion and deletion in
CRISPRs is slow enough such that they do not appear to change during an outbreak (Table 3.2).
CRISPRs may also reflect the specific phage and plasmid pool in the environment and hence
contain ecologically and geographically meaningful information for bacteria (32, 52). As a result,
CRISPRs may be useful for tracing an outbreak clone of Salmonella to the specific farm or food
processing plant which serves as the reservoir for the source strain of an outbreak. In conclusion,
the current MLST scheme effectively subtyped the two most common PFGE patterns of S.
Enteritidis and thus could enhance cluster detection and outbreak investigation capabilities. This
MLST method has the potential to be integrated into public surveillance laboratories to
complement PFGE for S. Enteritidis outbreak investigations.
77
It has been previously suggested that CRISPRs are poor epidemiological markers in
enterobacteria due to the slow rate of spacer acquisition (50). However, that study only analyzed
16 complete Salmonella genomes for CRISPRs, and only four of them were from the same
serovar as strains analyzed in the current study. Additionally, Touchon et al. only included in
their study one isolate of serovars Typhimurium, Enteritidis, Newport, and Heidelberg, so the true
value of CRISPRs for epidemiologic investigations could not be fully appreciated. Our study
analyzed 26, 34, 15 and 20 isolates from these serovars, respectively, and demonstrated that
CRISPR sequences may be implemented for epidemiologic investigations. We are currently
testing this hypothesis using larger numbers of isolates obtained from current and past Salmonella
outbreaks.
This MLST scheme has several other advantages that make it a potential subtyping
method for routine surveillance of Salmonella. First, the primers in this MLST scheme were
designed to have the same annealing temperature for all four markers so that it can be
conveniently performed in large-scale epidemiologic investigations. Second, the number of the
markers targeted was minimized to two virulence genes and two CRISPRs so that time and
expense can be saved during routine typing of Salmonella strains (37). Third, all four markers,
fimH, sseL, CRISPR1 and CRISPR2, are present in the major serovars of Salmonella and also in
all published genomes of Salmonella serovars, so the current MLST scheme is widely applicable.
Although this MLST scheme shows great promise, future research is needed to further validate it
for molecular epidemiologic purposes. For example, future research is needed using a random
collection of isolates representing a larger number of outbreaks, or otherwise epidemiologically
related isolates, to accurately compare the epidemiologic concordance of the present MLST
scheme with PFGE. In conclusion, the MLST scheme described in the current study maybe an
excellent subtyping method for tracking the farm-to-fork spread of the most prevalent serovars of
Salmonella during outbreaks.
78
3.6 Acknowledgements
We thank Dr. Bindhu Verghese for technical guidance throughout the study, especially
for the idea of combining CRISPRs into one allele to construct the cluster analysis. We also
acknowledge the Penn State Genomics Core Facility - University Park, PA for DNA sequencing.
This study was supported by a U.S. Department of Agriculture Special Milk Safety grant to the
Pennsylvania State University (contract: 2009-34163-20132).
79
3.7 References
1. Barrangou, R., C. Fremaux, H. Deveau, M. Richards, P. Boyaval, S. Moineau, D. A.
Romero, and P. Horvath. 2007. CRISPR provides acquired resistance against viruses in
prokaryotes. Science. 315:1709-1712.
2. Beranek, A., C. Mikula, P. Rabold, D. Arnhold, C. Berghold, I. Lederer, F. Allerberger,
and C. Kornschober. 2009. Multiple-locus variable-number tandem repeat analysis for
subtyping of Salmonella enterica subsp. enterica serovar Enteritidis. Int. J. Med. Microbiol.
299:43-51.
3. Brouns, S. J. J., M. M. Jore, M. Lundgren, E. R. Westra, R. J. H. Slijkhuis, A. P. L.
Snijders, M. J. Dickman, K. S. Makarova, E. V. Koonin, and J. van der Oost. 2008.
Small CRISPR RNAs guide antiviral defense in prokaryotes. Science. 321:960-964.
4. CDC. 2006. Salmonella Annual Summary 2005.
http://www.cdc.gov/ncidod/dbmd/phlisdata/salmtab/2006/SalmonellaAnnualSummary2006.p
df
5. CDC. 2007. Salmonella Annual Summary 2006.
http://www.cdc.gov/ncidod/dbmd/phlisdata/salmtab/2006/SalmonellaAnnualSummary2006.p
df
6. CDC. 2008. Outbreak of Salmonella serotype Saintpaul infections associated with multiple
raw produce items --- United States, 2008. MMWR Morb. Mortal. Wkly. Rep. 57:929-934.
7. CDC. 2008. OutbreakNet Foodborne Outbreak Online Database.
http://wwwn.cdc.gov/foodborneoutbreaks/
8. CDC. 2010. Investigation update: Multistate Outbreak of Human Salmonella Enteritidis
Infections Associated with Shell Eggs. http://www.cdc.gov/salmonella/enteritidis/
80
9. CDC. 2010. Investigation update: multistate outbreak of human Salmonella Montevideo
infections. http://www.cdc.gov/salmonella/montevideo/index.html
10. CDC. 2010. Preliminary FoodNet data on the incidence of infection with pathogens
transmitted commonly through food --- 10 States, 2009.
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5914a2.htm
11. Chen, Y., W. Zhang, and S. J. Knabel. 2007. Multi-virulence-locus sequence typing
identifies single nucleotide polymorphisms which differentiate epidemic clones and outbreak
strains of Listeria monocytogenes. J. Clin. Microbiol. 45:835-846.
12. Deveau, H., R. Barrangou, J. E. Garneau, J. Labonte, C. Fremaux, P. Boyaval, D. A.
Romero, P. Horvath, and S. Moineau. 2007. Phage response to CRISPR-encoded resistance
in Streptococcus thermophilus. J. Bacteriol. 190: 1390-1400.
13. Efrain, M. R., M. A. Fair, R. Gautom, D. N. Cameron, S. B. Hunter, B. Swaminathan,
and T. J. Barrett. 2006. Standardization of pulsed-field gel electrophoresis protocols for the
subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne
Pathog. Dis. 3:59-67.
14. Echeita, M. A., S. Herrera, and M. A. Usera. 2001. Atypical, fljB-negative Salmonella
enterica subsp. enterica strain of serovar 4,5,12:i:- appears to be a monophasic variant of
serovar Typhimurium. J. Clin. Microbiol. 39:2981-2983.
15. Endo, T., K. Ikeo, and T. Gojobori. 1996. Large-scale search for genes on which positive
selection may operate. Mol. Biol. Evol. 13:685-690.
16. Fakhr, M. K., L. K. Nolan, and C. M. Logue. 2005. Multilocus sequence typing lacks the
discriminatory ability of pulsed-field gel electrophoresis for typing Salmonella enterica
serovar Typhimurium. J. Clin. Microbiol. 43:2215-2219.
17. Foley, S. L., D. G. White, P. F. McDermott, R. D. Walker, B. Rhodes, P. J. Fedorka-
Cray, S. Simjee, and S. Zhao. 2006. Comparison of subtyping methods for differentiating
81
Salmonella enterica serovar Typhimurium isolates obtained from food animal sources. J. Clin.
Microbiol. 44:3569-3577.
18. Foley, S. L., S. Zhao, and R. D. Walker. 2007. Comparison of molecular typing methods
for the differentiation of Salmonella foodborne pathogens. Foodborne Pathog. Dis. 4:253-276.
19. Gerner-Smidt, P., K. Hise, J. Kincaid, S. Hunter, S. Rolando, E. Hyytiä-Trees, E. M.
Ribot, B. Swaminathan, and Pulsenet Taskforce. 2006. PulseNet USA: a five-year update.
Foodborne Pathog. Dis. 3:9-19.
20. Giammanco, G. M., C. Mammina, C. Romani, I. Luzzi, A. M. Dionisi, and A. Nastasi.
2007. Evaluation of a modified single-enzyme amplified fragment length polymorphism (SE-
AFLP) technique for subtyping Salmonella enterica serotype Enteritidis. Res. Microbiol.
158:10-17.
21. Grimont, P. A. D., and F. Weill. 2007. Antigenic formulae of the Salmonella serovars, 9th
ed. WHO Collaborating Centre for Reference and Research on Salmonella. Institut Pasteur,
Paris, France.
22. Grissa, I., and C. Drevet. 2010. CRISPRs web-service. http://crispr.u-psud.fr/crispr/
23. Grissa, I., G. Vergnaud, and C. Pourcel. 2008. CRISPRcompar: a website to compare
clustered regularly interspaced short palindromic repeats. Nucl. Acids Res. 36:W145-148.
24. Groenen, P. M., A. E. Bunschoten, D. Soolingen, and J. D. Errtbden. 1993. Nature of
DNA polymorphism in the direct repeat cluster of Mycobacterium tuberculosis; application
for strain differentiation by a novel typing method. Mol. Microbiol. 10:1057-1065.
25. Guard, J. 2010. Evolutionary trends in two strains of Salmonella enterica subsp. I serovar
Enteritidis PT13a that vary in virulence potential.
http://www.ncbi.nlm.nih.gov/genomes/static/Salmonella_SNPS.html
82
26. Hale, C. R., P. Zhao, S. Olson, M. O. Duff, B. R. Graveley, L. Wells, R. M. Terns, and M.
P. Terns. 2009. RNA-Guided RNA cleavage by a CRISPR RNA-Cas protein complex. Cell.
139:945-956.
27. Hanning B. Irene, J.D. Nutt, Steven C. Ricke. 2009. Salmonellosis outbreaks in the United
States due to fresh produce: sources and potential intervention measures. Foodborne Pathog.
Dis. 6:635-648.
28. Harbottle, H., D. G. White, P. F. McDermott, R. D. Walker, and S. Zhao. 2006.
Comparison of multilocus sequence typing, pulsed-field gel electrophoresis, and
antimicrobial susceptibility typing for characterization of Salmonella enterica serotype
Newport isolates. J. Clin. Microbiol. 44:2449-2457.
29. Hoe, N., K. Nakashima, D. Grigsby, X. Pan, S. J. Dou, S. Naidich, M. Garcia, E. Kahn,
D. Bergmire-Sweat, and J. M. Musser. 1999. Rapid molecular genetic subtyping of
serotype M1 group A Streptococcus strains. Emerg. Infect. Dis. 5:254-263.
30. Horvath, P., and R. Barrangou. 2010. CRISPR/Cas, the immune system of bacteria and
archaea. Science. 327:167-170.
31. Hunter, P. R., and M. A. Gaston. 1988. Numerical index of the discriminatory ability of
typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26:2465-
2466.
32. Kunin, V., S. He, F. Warnecke, B. S. Peterson, M. Haynes, N. Ivanova, L. L. Blackall, M.
Breitbart, F. Rohwer, D. Mcmahon, and P. Hugenholtz. 2008. A bacterial metapopulation
adapts locally to phage predation despite global dispersal. Genome Res. 18:293-297.
33. Lan, R., P. R. Reeves, and S. Octavia. 2009. Population structure, origins and evolution of
major Salmonella enterica clones. Infect. Genet. Evol. 9:996-1005.
34. Lindstedt, B. -., M. Torpdahl, E. M. Nielsen, T. Vardund, L. Aas, and G. Kapperud.
2007. Harmonization of the multiple-locus variable-number tandem repeat analysis method
83
between Denmark and Norway for typing Salmonella Typhimurium isolates and closer
examination of the VNTR loci. J. Appl. Microbiol. 102:728-735.
35. Lindstedt, B., E. Heir, E. Gjernes, and G. Kapperud. 2003. DNA fingerprinting of
Salmonella enterica subsp. enterica serovar typhimurium with emphasis on phage type
DT104 based on variable number of tandem repeat Loci. J. Clin. Microbiol. 41:1469-1479.
36. Lindstedt, B., E. Heir, T. Vardund, and G. Kapperud. 2000. Fluorescent amplified-
fragment length polymorphism genotyping of Salmonella enterica subsp. enterica serovars
and comparison with pulsed-field gel electrophoresis Typing. J. Clin. Microbiol. 38:1623-
1627.
37. Maiden, M. C. J. 2006. Multilocus sequence typing of bacteria. Annu. Rev. Microbiol.
60:561-588.
38. Pourcel, C., G. Salvignol, and G. Vergnaud. 2005. CRISPR elements in Yersinia pestis
acquire new repeats by preferential uptake of bacteriophage DNA, and provide additional
tools for evolutionary studies. Microbiology. 151:653-663.
39. Ross, I. L., and M. W. Heuzenroeder. 2005. Use of AFLP and PFGE to discriminate
between Salmonella enterica serovar Typhimurium DT126 isolates from separate food-
related outbreaks in Australia. Epidemiol. Infect. 133:635-644.
40. Ross, I. L., and M. W. Heuzenroeder. 2009. A comparison of two PCR-based typing
methods with pulsed-field gel electrophoresis in Salmonella enterica serovar Enteritidis. Int. J.
Med. Microbiol. 299:410-420.
41. Scaria, J., R. U. M. Palaniappan, D. Chiu, J. A. Phan, L. Ponnala, P. McDonough, Y. T.
Grohn, S. Porwollik, M. McClelland, C. Chiou, C. Chu, and Y. Chang. 2008. Microarray
for molecular typing of Salmonella enterica serovars. Mol. Cell. Probes. 22:238-243.
42. Schouls, L. M., S. Reulen, B. Duim, J. A. Wagenaar, R. J. L. Willems, K. E. Dingle, F. M.
Colles, and J. D. A. Van Embden. 2003. Comparative genotyping of Campylobacter jejuni
84
by amplified fragment length polymorphism, multilocus sequence typing, and short repeat
sequencing: strain diversity, host range, and recombination. J. Clin. Microbiol. 41:15-26.
43. Sorek, R., V. Kunin, and P. Hugenholtz. 2008. CRISPR — a widespread system that
provides acquired resistance against phage in bacteria and archaea. Nat. Rev. Microbiol.
6:181-186.
44. Struelens, M. J. 1996. Consensus guidelines for appropriate use and evaluation of microbial
epidemiologic typing systems. Clin. Microbiol. Infect. 2:2-11.
45. Tamada, Y., Y. Nakaoka, K. Nishimori, A. Doi, T. Kumaki, N. Uemura, K. Tanaka, S.
Makino, T. Sameshima, M. Akiba, M. Nakazawa, and I. Uchida. 2001. Molecular typing
and epidemiological study of Salmonella enterica serotype Typhimurium isolates from cattle
by fluorescent amplified-fragment length polymorphism fingerprinting and Pulsed-Field Gel
Electrophoresis. J. Clin. Microbiol. 39:1057-1066.
46. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: molecular evolutionary
genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24:1596-1599.
47. Tankouo-Sandjong, B., A. Sessitsch, E. Liebana, C. Kornschober, F. Allerberger, H.
Hächler, and L. Bodrossy. 2006. MLST-v, multilocus sequence typing based on virulence
genes, for molecular typing of Salmonella enterica subsp. enterica serovars. J. Microbiol.
Methods. 69:23-36.
48. Tankouo-Sandjong, B., A. Sessitsch, N. Stralis-Pavese, E. Liebana, C. Kornschober, F.
Allerberger, H. Hächler, and L. Bodrossy. 2008. Development of an oligonucleotide
microarray method for Salmonella serotyping. Microb Biotechnol. 1:513-522.
49. Torpdahl, M., M. N. Skov, D. Sandvang, and D. L. Baggesen. 2005. Genotypic
characterization of Salmonella by multilocus sequence typing, pulsed-field gel
electrophoresis and amplified fragment length polymorphism. J. Microbiol. Methods. 63:173-
184.
85
50. Touchon, M., and P. C. E. Rocha. 2010. The small, slow and specialized CRISPR and anti-
CRISPR of Escherichia and Salmonella. PLoS One. 5:e11126.
51. USDA-ERS. 2009. Foodborne illness cost calculator: Salmonella.
http://www.ers.usda.gov/Data/FoodborneIllness/salm_Intro.asp.
52. Vale, P. F., and T. J. Little. 2010. CRISPR-mediated phage resistance and the ghost of
coevolution past. Proc. R. Soc. Lond., B, Biol. Sci. 277: 2097-2103.
53. Voetsch, A., T. Van Gilder, F. Angulo, M. Farley, S. Shallow, R. Marcus, P. Cieslak, V.
Deneen, and R. Tauxe. 2004. FoodNet estimate of the burden of illness caused by
nontyphoidal Salmonella infections in the United States. Clin. Infect. Dis. 38:127-134.
54. Zheng, J., C. E. Keys, S. Zhao, J. Meng, and E. W. Brown. 2007. Enhanced subtyping
scheme for Salmonella Enteritidis. Emerging Infect. Dis.13:1932-1935.
86
Chapter 4 Characterization of clinical, poultry and environmental Salmonella
Enteritidis isolates using multilocus sequence typing based on virulence genes
and CRISPRs
Fenyun Liu1, Subhashinie Kariyawasam
2, Bhushan M. Jayarao
2, 3, Rodolphe Barrangou
4,
Peter Gerner-Smidt5, Efrain Ribot
5, Edward G. Dudley
1 and Stephen J. Knabel
1*
1Department of Food Science, the Pennsylvania State University, University Park,
Pennsylvania 16802;
2Animal Diagnostic Laboratory, Orchard Drive, the Pennsylvania State University,
University Park, Pennsylvania 16802;
3Department of Veterinary and Biomedical Sciences, the Pennsylvania State University,
University Park, Pennsylvania 16802;
4Danisco USA Incorporation, 3329 Agriculture Drive, Madison, Wisconsin 53716;
5Centers for Disease Control and Prevention, Atlanta, Georgia 30333
*Corresponding author. Mailing address: 437 Food Science Building, The Pennsylvania
State University, University Park, PA 16802, US. Phone: 814-863-1372. Email:
87
4.1 Abstract
Salmonella enterica subsp. enterica serovar Enteritidis has consistently been a major
cause of foodborne salmonellosis in the United States. Two major food vehicles for S. Enteritidis
are contaminated eggs and chicken. Improved subtyping methods are needed to accurately track
specific strains of S. Enteritidis related to human salmonellosis throughout the poultry and egg
food system. A multilocus sequence typing (MLST) scheme based on virulence genes (fimH and
sseL) and CRISPRs (Clustered Regularly Interspaced Short Palindromic Repeats) was developed
and used to characterize 34 clinical isolates, 70 poultry isolates and 63 hen house environment
isolates of S. Enteritidis. A total of 27 sequence types (STs) were identified for the 167 S.
Enteritidis isolates. The MLST scheme identified four persistent and predominate STs circulating
among U.S. clinical isolates, and poultry and hen house environments in Pennsylvania. It also
identified a potential environment-specific sequence type. Moreover, cluster analysis based on
fimH and sseL identified three epidemic clones and one outbreak clone of S. Enteritidis, as well as
11 singletons. Significant differences in virulence gene sequences between singletons and the
other STs suggested that singletons might have different virulence capacity than other STs. The
MLST scheme may provide information about the ecological origin of S. Enteriditis isolates,
potentially identifying strains that differ in virulence capacity.
88
4.2 Introduction
In the United States, Salmonella is the leading cause of bacterial foodborne disease, with
approximately 1.4 million human cases each year since 1996 (41). The second most-reported
serovar of Salmonella for human diseases is Salmonella enterica subspecies enterica serovar
Enteritidis (S. Enteritidis), which causes nearly as many human cases as S. Typhimurium, the
most prevalent serovar (8). The major food vehicle for S. Enteritidis is shell eggs and 80% of the
S. Enteritidis outbreaks and approximately 50,000 to 110,000 cases are egg-associated in the U.S.
each year (6, 34). The most recent S. Enteritidis outbreak was associated with eggs, in which
1,519 people got infected (9). Another common food vehicle is chicken, consumption of which is
considered as another risk factor for S. Enteritidis infections in humans (22, 30).
The chicken and egg food system is complex and contains a large number of niches that
may be potential sources of S. Enteritidis (Fig. 4.1) (21). S. Enteritidis has been isolated from a
wide variety of animals, such as rodents, wild birds and insects, which could serve as potential
reservoirs (17). Additional potential reservoirs for S. Enteritidis include chicken manure, sewage
and other moist and organic materials in farm environments (6). Furthermore, oral S. Enteritidis
infection in poultry could occur via contaminated water and feed (6). Infection of S. Enteritidis
among chickens can spread rapidly by direct contact with infected birds or with contaminated
materials within densely populated poultry houses (6, 17). Additionally, eggs can become
contaminated internally when the ovaries of layers are colonized by S. Enteritidis; this process is
called vertical transmission (17, 31). Eggs can also be contaminated externally by feces and
environments containing S. Enteritidis; this is referred to as horizontal transmission (14). In order
to control S. Enteritidis in poultry, one of the interventions employed on farms is egg quality
assurance programs, which involve acquisition of S. Enteritidis free chicks, control of pests
89
(including rodents and insects), use of S. Enteritidis-free feeds, and routine microbiologic testing
for S. Enteritidis in the farm environment (6).
Characterization of S. Enteritidis isolates in different niches on chicken farms and from
patients can help track dissemination of specific strains related to human salmonellosis and thus
identify reservoirs and routes of transmission. Before considering the epidemiology of
Salmonella, it is important to first define epidemic clone (EC), and outbreak clone (OC).
Epidemic clone is a strain or group of strains descended asexually from a single ancestral cell
(source strain) that is involved in one epidemic, and can often include several outbreaks (11).
Outbreak clone is a strain or group of strains descended asexually from a single ancestral cell
(source strain) that is involved in one outbreak (11).
Several molecular subtyping methods have been developed to characterize strains and
study the epidemiology of S. Enteritidis, including amplified fragment length polymorphism
(AFLP) (27, 15, 19, 36, 38), multiple loci variable number tandem repeat analysis (MLVA) (3, 5,
12, 35), and pulsed-field gel electrophoresis (PFGE) (18). PFGE is currently the ―gold standard‖
method used by public health surveillance laboratories for tracking foodborne pathogens
including Salmonella (18). The main advantage of PFGE is its high discriminatory power (i.e.
ability to separate unrelated strains) for subtyping most serovars of Salmonella. However, PFGE
lacks discriminatory power for highly clonal serovars like S. Enteritidis (18, 42). For example,
the most recent S. Enteritidis outbreak due to shell eggs was caused by the most common PFGE
pattern for S. Enteritidis in the PulseNet database and thus not all isolates may be related to this
outbreak (9). The lack of adequate discriminatory power makes it difficult to track a specific
strain of S. Enteritidis in the food system. Besides occasional inadequate discriminatory power,
PFGE does not provide appropriate information to infer phylogenetic relationships among
subtypes (33). Another subtyping method, MLVA, was reported to have higher discriminatory
power than PFGE for S. Enteritidis (3, 5, 12, 35). However, in some circumstances, strains that
90
had the same MLVA type were separated by PFGE (5). Moreover, replicates of the same strains
of Salmonella have been shown to have different number of repeat units at a specific locus (7, 13),
which makes accurate interpretation of results difficult.
Compared to PFGE and MLVA, Multilocus Sequence Typing (MLST), which targets
nucleotide sequence differences in several DNA loci, generates discreet, highly informative,
highly portable and reproducible data. Moreover, MLST is a well-accepted tool for studying the
population structure, evolution and diversity of bacteria (25). Recently, a new MLST scheme
based on virulence genes (fimH and sseL) and CRISPRs (Clustered Regularly Interspaced
Palindromic Repeats) was shown to provide better separation of S. Enteritidis than PFGE (29).
CRISPRs encode tandem sequences containing 21-47 bp DRs (direct repeats) and spacers of
similar size (Fig. 4.2). Spacers are short DNA sequences obtained from foreign nucleic acids
such as phages or plasmids and are inserted into bacterial chromosome to protect them from
infection by homologous phages and plasmids (2). Therefore, CRISPRs reflect the specific phage
and plasmid pools in the environment and hence contain ecological and geographic information
of the bacteria present there (24, 40). As a result, CRISPRs might be useful for tracing back a
clone of S. Enteritidis to the specific niches in the farm or food processing plant where it
originated. Therefore, the objective of the present study was to characterize S. Enteritidis isolates
from different sources using this scheme to investigate its epidemiology. With a better
understanding of the epidemiology of S. Enteritidis, more effective intervention strategies can be
established to prevent future S. Enteritidis outbreaks due to eggs and poultry meat.
91
4.3 Materials and methods
Bacterial isolates and DNA extraction. A summary of all Salmonella Enteritidis
isolates used in this study are listed in Table 4.1. A total of 167 isolates were obtained as follows:
34 from Centers for Disease Control and Prevention (CDC), 86 from the Pennsylvania Egg
Quality Assurance Program (PEQAP) and 47 from the Animal Diagnostic Lab (ADL) at the
Pennsylvania State University (Table S2). All 34 clinical isolates were previously analyzed (29),
among which 32 isolates were collected from 11 S. Enteritidis outbreaks and the other 2 isolates
were sporadic isolates. Bacterial isolates were stored at -80°C in 20% glycerol. When needed,
isolates were grown overnight in Tryptic Soy Broth (TSB) (Difco Laboratories, Becton Dickinson,
Sparks, MD) at 37°C. For all isolates, DNA was extracted using the UltraClean Microbial DNA
extraction kit (Mo Bio Laboratories, Solana Beach, CA) and stored at -20°C before use.
PCR amplification. Primers for all four markers were designed based upon consensus
alignments of the published S. Typhimurium LT2 genome (accession number AE006468) using
Primer 3.0 (http://frodo.wi.mit.edu/primer3/) (Table 4.2). PCR amplifications were performed
using a Taq PCR master mix kit (Qiagen Inc., Balencia, CA) in a Mastercycler PCR thermocycler
(Eppendorf Scientific, Hamburg, Germany). A 25 µl PCR reaction system contained 12.5 µl Taq
PCR master mix, 9.5 µl PCR-grade water, 1.0 µl DNA template, 1.0 µl forward primer (final
concentration, 0.4 µM) and 1.0 µl reverse primer (final concentration, 0.4 µM). A single PCR
cycling condition was used to amplify all four markers (initial denaturation at 94 °C for 10 min;
28 cycles of 94°C for 1 min, 55°C for 1 min, 72°C for 1 min; final extension at 72°C for 10 min).
DNA sequencing. After PCR, products for sequencing were treated by adding 1/20
volume of shrimp alkaline phosphatase (1 U/µl, USB Corp. Cleveland, OH) and 1/20 volume of
exonuclease I (10 U/µl, USB Corp). The mixture was then incubated at 37°C for 45 min to
92
degrade the primers and unincorporated dNTPs. After that, the mixture was incubated at 80°C
for 15 min to inactivate the enzymes. PCR products were then sent to the Genomics Core Facility
at the Pennsylvania State University for sequencing using the ABI Data 3730XL DNA Analyzer.
In order to obtain complete DNA sequences of fimH and sseL, two more primers targeting the
internal regions of these two genes were used together with the forward and reverse primers
(Table 4.2). Both DNA strands of the amplicons were sequenced.
Sequence analysis and sequence type assignment. For fimH and sseL, sequences were
aligned and single nucleotide polymorphisms (SNPs) were identified using MEGA 4.0 (37). For
CRISPR1 and CRISPR2, analyses of the spacer arrangements were performed using
CRISPRcompar (20) and spacers were visualized as described by Deveau et al. (16). Different
allelic types (ATs) (sequences with at least one-nucleotide difference or one-spacer difference in
the case of CRISPRs) were assigned arbitrary numbers. The combination of 4 alleles (fimH, sseL,
CRISPR1 and CRISPR2) determined its allelic profile and each unique allelic profile was
designated as a sequence type (ST).
Cluster analysis. Cluster analyses were performed based on allelic profile data by the
unweighted pair group method with arithmetic mean (UPGMA) and results were visualized using
the tree drawing tool on PubMLST (www.pubmlst.org). CRISPR1 and CRISPR2 were combined
into one allele for a more accurate cluster analysis, because CRISPR1 and CRISPR2 are likely to
be spatially linked (39).
Nucleotide sequence accession number. DNA sequences of the four genetic MLST
markers were deposited in GenBank under accession numbers HQ329919 to HQ329971.
93
4.4 Results
Results of MLST and sequence type distribution. In order to gain insights into the
sources and routes of transmission of S. Enteritidis contamination, the 167 isolates were
characterized using an MLST scheme based on virulence genes and CRISPRs which was
previously developed in our laboratories (Dr. Stephen Knabel and Dr. Edward Dudley). fimH,
sseL, CRISPR1 and CRISPR2 identified 12, 13, 14 and 20 allelic types, respectively. In total, 27
sequence types (STs) were identified for all 167 isolates (Fig. 4.6 and Table S2). There were 9,
12 and 15 STs in clinical, poultry and environmental isolates, respectively. For clinical isolates,
the 9 STs were E ST1, 2, 3, 4, 5, 6, 7, 8 and 9. The number of clinical isolates in each sequence
type is listed in Fig. 4.6. Out of the 9 STs found in clinical isolates, 5 STs (E ST2, 5, 6, 7 and 9)
were not found in either poultry or environmental isolates (Fig. 4.6). Those 5 STs came from
California, Georgia, Maine, Michigan and Ohio. For poultry isolates, the 12 STs included E ST1,
3, 4, 8, 12, 15, 21, 22, 23, 24, 25 and 26, and 7 of them (E ST15, 21, 22, 23, 24, 25 and 26) were
only found in poultry isolates. For the 15 STs in environmental isolates (E ST1, 3, 4, 8, 10, 11,
12, 13, 14, 16, 17, 18, 19, 20 and 27), 10 STs (E ST10, 11, 13, 14, 16, 17, 18, 19, 20 and 27) were
exclusively found in environment.
Predominant STs. An uneven distribution of the 27 STs was observed between different
sources. Overall, the 5 STs designated E ST1, 3, 4, 8 and 10 accounted for 19 %, 17%, 25%, 8%
and 7% of the total isolates, respectively, accounted for 76% of all isolates. Out of the 5
predominant STs, 4 of them (E ST1, 3, 4, and 8) were found in clinical, poultry and
environmental isolates. E ST1 made up 12% of clinical isolates, 33% of poultry isolates and 8%
of environmental isolates, respectively. E ST3 was found in 15% of clinical, 9% of poultry and
2% of environmental isolates. E ST4 accounted for 15% of clinical isolates, 30% of poultry
94
isolates and 3 % environmental isolates. E ST8 comprised 21% of clinical isolates, 14% of
poultry isolates and 40% of environmental isolates. E ST10 was only found in environmental
isolates, where it comprised 21% of all environmental isolates.
Cluster analyses. A cluster diagram based on virulence genes identified three epidemic
clones (ECI, ECII and ECIII), which included STs from multiple outbreaks, were identified by
fimH and sseL (Fig. 4.4). ECI contained 9 STs, E ST3, 4, 5, 8, 10, 12, 14, 18 and 27. In total,
110 (66% of total isolates) belonged to ECI, which was the largest cluster and contained 18
clinical, 41 poultry and 51 environmental isolates. ECII contained 22% of all isolates (8 clinical,
24 poultry, and 5 environmental isolates) and 3 STs (E ST1, 9 and 26). ECIII contained 3 STs (E
ST2, 7 and 13) which included 6 clinical isolates and 1 environmental isolate. One outbreak
clone (OC), E ST6, contained 2 clinical isolates from the same outbreak. Besides the 3 ECs and 1
OC, 11 singletons occupied a single branch on the tree. Among the 11 singletons, 5 STs (E ST11,
16, 17, 19 and 20) were found in the farm environment and 6 (E ST15, 21, 22, 23, 24 and 25)
were found in poultry isolates. These 6 poultry singletons were either sampled from eggs in
broiler hatcheries with hatchability problems or necropsy isolates from sick broilers.
Incorporation of CRISPRs into the MLST method separated isolates within the 3 ECs
(Fig. 4.5). Among the 15 STs in the 3 ECs, 4 STs (E ST1, 3, 4 and 8) were found in all sources
(clinical, poultry and environmental). These STs were also the predominant STs among all
isolates (Fig. 4.3). E ST12 was found in both poultry and environment. The other 10 STs were
from a single source including 4 STs (E ST1, 2, 5 and 9) found only in clinical isolates, 2 STs (E
ST26 and 27) found only in poultry isolates and 4 STs (E ST10, 13, 14 and 18) found only in
environmental isolates.
Spacer arrangements among STs. Fig. 4.6 shows the differences in spacer
arrangements among STs in CRISPR1 and CRISPR2. In CRISPR1, the number of spacers
ranged from 2 to 25; for CRISPR2, the number of spacers ranged from 3 to 25. Generally, there
95
were great similarities among STs in the 3 ECs and the OC. The singleton E ST16 also shared
spacers with STs in the 3 ECs and the OC; however many other spacers were deleted. The other
10 singletons contained totally different spacers from each other, except E ST21 and E ST22,
which shared most spacers within CRISPR1, and had identical CRISPR2 loci.
96
Table 4.1. Sources, sample types and isolation information for the 167 S. Enteritidis isolates
analyzed in the present study
Sources No. of
isolates Samples States
1 Source
2 Year
Clinical 34 Clinical (stool; foods related to
outbreaks)
13
states CDC
2001-
2009
Poultry 70
Poultry: 46 egg and necropsy
isolates of broiler PA ADL
2007-
2009
Poultry: 3 egg isolates from layer
houses PA PEQAP
1998-
1999
Poultry: 21 egg isolates from layer
houses PA PEQAP
2007-
2010
Environment 63
Environmental: 46 drag swabs PA PEQAP 1998-
1999
Environmental: 17 drag swabs PA PEQAP;
ADL
2007-
2010
1Clinical isolates were from 13 states, including CA, CO, CT, GA, ID, ME, MI, MN, OH, OR,
PA, SC and WV.
2 Isolates were received from CDC (Centers for Disease Control and Prevention), PEQAP
(Pennsylvania Egg Quality Assurance Program) and ADL (Animal Diagnostic Lab) in
Pennsylvania State University.
97
Table 4.2. Primers used to amplify and sequence the four MLST markers
Marker Primer sequence (5'-3') Note
fimH CGTCGTCATAAAAGGAAAAA Forward primer for both amplification and sequencing
GAACAAAACACAACCAATAGC Reverse primer for both amplification and sequencing
CTCGCCAGACAATGTTTACT Reverse primer for sequencing internal region
CATTCACTTCGCAGTTTTG Forward primer for sequencing internal region
sseL AGGAAACAGAGCAAAATGAA Forward primer for both amplification and sequencing
TAAATTCTTCGCAGAGCATC Reverse primer for both amplification and sequencing
GGAGTTGAAAATCTTTGGTG Reverse primer for sequencing internal region
TTTACCGAGAGAAAAGGTGA Forward primer for sequencing internal region
CRISPR1 GATGTAGTGCGGATAATGCT Forward primer for both amplification and sequencing
GGTTTCTTTTCTTCCTGTTG Reverse primer for both amplification and sequencing
CRISPR2 ACCAGCCATTACTGGTACAC Forward primer for both amplification and sequencing
ATTGTTGCGATTATGTTGGT Reverse primer for both amplification and sequencing
98
Figure 4.1. Potential routes of transmission of S. Enteritidis contamination throughout the egg
food system.
99
Figure 4.2. Schematic view of the two CRISPR systems in Salmonella Enteritidis strain P125109.
Direct repeats and spacers are represented by black diamonds and white rectangles, respectively.
The terminal direct repeats are represented by white diamonds. Numbers of direct repeats and
spacers are represented by the numbers of diamonds and white rectangles, respectively. L stands
for leader sequence. cas genes are in grey while other core flanking genes (ygcF, iap and ptps)
are in white. Primer targeting sites are indicated by upward ponting arrows. The figure is not
drawn to scale.
CRISPR1
CRISPR2
5
’
5
’
3
’
3
’
100
Figure 4.3. Frequency of the five predominant sequence types (E ST1, 3, 4, 8 and 10) in clinical,
poultry and environmental isolates.
The five predominant sequence types accounted for 76% of all isolates analyzed in the present
study. All unlabeled pie slices in Fig. (b), (c), (d) are STs unique to that given source, except the
pie slice of E ST10.
(a)
(b) (c) (d)
101
Figure 4.4. Cluster diagram based on only fimH and sseL for all 27 sequence types.
MLST identified three epidemic clones (ECI, ECII and ECIII) and one outbreak clone. Clinical,
poultry and environmental isolates are designated by c, p and e, respectively. The number of
isolates from each source is indicated before the source designation.
102
Figure 4.5. Cluster diagram based on virulence genes and CRISPRs for all 27 sequence types.
Clinical, poultry and environmental isolates are designated by c, p and e, respectively. The
number of isolates from each source is indicated before the source designation.
103
ST Source Cluster CRISPR1 CRISPR2
1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25 1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
P125109 p p p p p p p p p p p p p p p p p
EST3 c(5), p(6), e(1) ECI p p p p p p p p p p p p p p p
EST10 e(13) ECI p p p p p p p p p p p p p p
EST2 c(5) ECIII p p p p p p p p p p p p p p p
EST13 e(1) ECIII p p p p p p p p p p p p p p p p
EST7 c(1) ECIII p p p p p p p p p p p p p p p p
EST9 c(4) ECII p p p p p p p p p p p p p p p p
EST12 p(3), e(5) ECI p p p p p p p p p p p p p p p p
EST1 c(4), p(23), e(5) ECII p p p p p p p p p p p p p p p p p
EST8 c(7), p(10), e(26) ECI p p p p p p p p p p p p p p p p p
EST6 c(2) OC p p p p p p p p p p p p p p p p p
EST14 e(3) ECI p p p p p p p p p p p p p p p p p
EST26 p(1) ECII p p p p p p p p p p p p p p p p p p
EST27 e(1) ECI p p p p p p p p p p p p p p p p p p
EST4 c(5), p(21), e2) ECI p p p p p p p p p p p p p p p p p p p
EST18 e(2) ECI p p p p p p p p p p p p p p p p p p
EST5 c(1) ECI p p p p p p p p p p p p p p p p
EST16 e(1) Singleton p p p p p
EST21 p(1) Singleton p p p p p p p p p p p p p p p p p p p p p p p p
EST22 p(1) Singleton p p p p p p p p p p p p p p p p p p p p p p p
EST17 e(1) Singleton p p p p p p p p p p l p p p p p p p p p p p p p p p p p p p p p p p p
EST20 e(1) Singleton p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
EST11 e(1) Singleton p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
EST23 p(1) Singleton p p p p p p p p p p p p p p p p p p p p p p
EST15 p(1) Singleton p p p p p p p p p p p p p p p p p p p p v p p p p p p p p p p p p p p p p
EST24 p(1) Singleton p p p p p p p p p p p p p p p p p p p p p p p
EST25 p(1) Singleton p p p p p p p p p p p p p p p p p p p
EST19 e(1) Singleton p p p p p p p p p p
Figure 4.6. Graphic representation of spacer arrangements in CRISPR1 and CRISPR2 of the 27 S.
Enteritidis sequence types.
Clinical, poultry and environmental isolates are designated by c, p and e, respectively. The
number of isolates from each source is indicated after the source designation. Repeats are not
included; only spacers are listed. The direction of the spacer arrangements is from the 5’ to 3’
leader end. Each unique spacer is represented by a unique combination of the background color
and the color of a particular character. Spacers were aligned and gaps represent the absent of a
particular spacer. Singletons had very distinct arrays of spacers and thus could not aligned.
P125109 is the strain of S. Enteritidis that has a whole genome sequence deposited in GenBank.
104
4.5 Discussion
S. Enteritidis outbreaks due to consumption of eggs and chicken pose a great public-
health and financial burden (6, 22, 30, 34). To prevent future outbreaks, it is important to identify
the reservoirs, sources and routes of transmission of S. Enteritidis throughout the chicken and egg
food system. Because S. Enteritidis is a highly clonal serovar, PFGE is often unable to accurately
differentiate outbreak clones (9, 42). Likewise, many other subtyping methods, including
ribotyping and plasmid profiling, were also not discriminatory enough differentiate S. Enteritidis
outbreak clones and thus were not useful tracking tools (26). Recently an MLST scheme based
on virulence genes and CRISPRs was shown to provide better discrimination for S. Enteritidis
than PFGE and may be a potential tool for tracking S. Enteritidis (29). To test its ability to track
specific subtypes of S. Enteritidis and to enhance our understanding of the epidemiology of S.
Enteritidis, a total of 167 S. Enteritidis isolates were subtyped by this MLST scheme and 27 STs
were identified (Fig. 6). Nine STs were observed among clinical isolates and 5 STs (E ST2, 5, 6,
7 and 9) were only found in clinical isolates (Fig. 6). It should be noted that our clinical isolates
were from various geographical locations within the U.S.; however, the poultry and
environmental isolates were all from Pennsylvania. Therefore, these STs may represent strains
that were not common in the PA poultry environment. In contrast, the other 4 STs (E ST1, 3, 4
and 8) were found in poultry and environmental isolates in PA (Fig. 6). For example, E ST4 was
found in 3 clinical isolates, 21 poultry and 2 environmental isolates from PA (Fig. 6 and S1).
This suggests that PA hen houses might be one of the sources of these clinical STs and thus the
MLST scheme based on virulence genes and CRISPRs might be used to track specific clinical
STs back to their geographic origin.
105
The present MLST scheme identified 5 predominant STs among all 167 isolates included
in the present study (Fig. 3). Because four STs (E ST1, 3, 4 and 8) were found in all sources
(clinical, poultry and environmental), they might represent four predominate STs circulating
among humans in the U.S. and poultry and hen house environments in PA. Moreover, these 4
STs were isolated over duration of ten years (Table 1 and S1), so they are also persistent STs
among humans in the U.S. and poultry industries in PA. As we were unable to find E ST10
among chicken and clinical isolates, this ST might represent an environment-adapted clone of S.
Enteritidis. A previous MLST study observed clustering of environmental versus clinical isolates
and suggested that environmental isolates might have decreased virulence to humans (23).
Likewise, we speculate that E ST10 might have decreased virulence as well, and plan to test this
hypothesis in the future.
In the present study, virulence genes were used to study the molecular epidemiology of S.
Enteritidis and identified 3 ECs and 1 OC (Fig. 4). Virulence genes had been previously used to
determine epidemic clones of Listeria monocytogenes (10) and also the epidemiology of
Salmonella among cow and human isolates, because they reflect virulence capacities of bacterial
isolates (1). It should be noted here that E ST6 might belong to an epidemic clone; however, no
other outbreak was identified in this cluster, probably due to the limited isolates included in the
present study, and thus it was designated an OC. Among the 3 ECs, ECI was the largest cluster
with 4 of the 5 predominant STs (E ST3, 4, 8 and 10 ) and 110 isolates (66%) from all 3 sources
(clinical, poultry and environmental) (Fig. 4). Therefore, ECI might represent a major epidemic
clone and may be responsible for causing most S. Enteritidis cases in the U.S. ECII might be
another major epidemic clone with the predominant sequence type E ST1. Isolates from all
sources were also found in ECII (Fig. 4). The existence of major clones of S. Enteritidis in
human and poultry was also observed by previous studies using pulsed-field gel electrophoresis
(PFGE) (4, 32). The 11 singletons identified in the present study were distant from the 3 ECs and
106
the OC on the cluster diagram and their virulence gene sequences varied significantly from other
STs (Fig. 4).
CRISPRs also separated singletons out from the 3 ECs and the OC due to dramatic
differences in spacer arrangements in both CRISPRs (Tables 10 and 11). Studies previously
demonstrated that bacteria from distant geographic space had strikingly different spacer
arrangements, most likely due to host-phage coevolution (24). Therefore, spacer arrangements
are a good indicator of bacterial adaptation to different microenvironments, which may be driven
by phages unique to specific niches. Therefore, singletons may represent unique ecotypes that are
distinct from STs in the 3 ECs and the OC. Six singletons (E ST15, 21, 22, 23, 24 and 25) may
be adapted to poultry and only pathogenic to chickens for the following reasons: 1) they were
only found in eggs with hatchability problems or necropsy isolates from sick broilers; 2) their
spacer arrangements were dramatically different from those in 3ECs and OC with no spacers
shared between them (Fig. 6); 3) and they showed significant differences in virulence gene
sequences as compared to those in 3ECs and the OC (Fig. 4). The other 5 singletons (E ST11, 16,
17, 19 and 20) were only found in the production environment, not in human and poultry, hence
they may be adapted to this environment. Spacers and virulence genes in these 5 singletons
differed significantly from those in ECs and the OC and thus they might be non pathogenic to
both humans and chickens. Future experiments are planned to compare the virulence of
singletons and clinical isolates. Host-phage coevolution may drive CRISPRs to evolve much
faster than virulence genes because ECs grouped by virulence genes were further separated into
many STs by CRISPRs (Fig. 4 and 5). This increased discrimination by CRISPRs was very
useful because they appeared to accurately separate outbreak clones within epidemic clones.
In conclusion, the present MLST scheme may be a valuable molecular subtyping method
for tracking the spread of S. Enteritidis throughout the poultry and egg food systems. Additional
research is needed to determine the source of S. Enteritidis contamination and the routes of
107
transmission using poultry isolates from other states in the U.S., as well as isolates from breeder
flocks.
108
4.6 Acknowledgements
We thank Dr. Bindhu Verghese for technical guidance throughout the study, especially
for the idea of combining CRISPRs into one allele in the cluster analysis. We also acknowledge
the Penn State Genomics Core Facility - University Park, PA for DNA sequencing. This study
was supported by a U.S. Department of Agriculture Special Milk Safety grant to the Pennsylvania
State University (contract: 2009-34163-20132).
109
4.7 References
1. Alcaine, S. D., Y. Soyer, L. D. Warnick, W. -. Su, S. Sukhnanand, J. Richards, E. D.
Fortes, P. McDonough, T. P. Root, N. B. Dumas, Y. Grohn, and M. Wiedmann. 2006.
Multilocus sequence typing supports the hypothesis that cow- and human-associated
Salmonella isolates represent distinct and overlapping populations. Appl. Environ. Microbiol.
72:7575-7585.
2. Barrangou, R., C. Fremaux, H. Deveau, M. Richards, P. Boyaval, S. Moineau, D. A.
Romero, and P. Horvath. 2007. CRISPR provides acquired resistance against viruses in
prokaryotes. Science. 315:1709-1712.
3. Beranek, A., C. Mikula, P. Rabold, D. Arnhold, C. Berghold, I. Lederer, F. Allerberger,
and C. Kornschober. 2009. Multiple-locus variable-number tandem repeat analysis for
subtyping of Salmonella enterica subsp. enterica serovar Enteritidis. Int. J. Med. Microbiol.
299:43-51.
4. Boonmar, S., A. Bangtrakulnonth, S. Pornrunangwong, J. Terajima, H. Watanabe, K.
Kaneko, and M. Ogawa. 1998. Epidemiological analysis of Salmonella Enteritidis isolates
from humans and broiler chickens in Thailand by phage typing and pulsed-field gel
electrophoresis. J. Clin. Microbiol. 36:971-974.
5. Boxrud, D., K. Pederson-Gulrud, J. Wotton, C. Medus, E. Lyszkowicz, J. Besser, and J.
M. Bartkus. 2007. Comparison of multiple-locus variable-number tandem repeat analysis,
pulsed-field gel electrophoresis, and phage typing for subtype analysis of Salmonella enterica
serotype Enteritidis. J. Clin. Microbiol. 45:536-543.
6. Braden, C. R. 2006. Salmonella enterica serotype Enteritidis and eggs: a national epidemic
in the United States. Clin. Infect. Dis. 43:512-517.
110
7. Call, D., L. Orfe, M. Davis, S. Lafrentz, and M. Kang. 2008. Impact of compounding error
on strategies for subtyping pathogenic bacteria. Foodborne Pathog. Dis. 5:505-516.
8. CDC. 2007. Salmonella Annual Summary 2006.
http://www.cdc.gov/ncidod/dbmd/phlisdata/salmtab/2006/SalmonellaAnnualSummary2006.p
df
9. CDC. 2010. Investigation update: Multistate Outbreak of Human Salmonella Enteritidis
Infections Associated with Shell Eggs. http://www.cdc.gov/salmonella/enteritidis/
10. Chen, Y., W. Zhang, and S. J. Knabel. 2007. Multi-virulence-locus sequence typing
identifies single nucleotide polymorphisms which differentiate epidemic clones and outbreak
strains of Listeria monocytogenes. J. Clin. Microbiol. 45:835-846.
11. Chen, Y., and S. Knabel. 2008. Strain typing, p. 203-239. In D. Liu (ed.), Handbook of
Listeria monocytogenes.
12. Cho, S., D. J. Boxrud, J. M. Bartkus, T. S. Whittam, and M. Saeed. 2007. Multiple-locus
variable-number tandem repeat analysis of Salmonella Enteritidis isolates from human and
non-human sources using a single multiplex PCR. FEMS Microbiol. Lett. 275:16-23.
13. Davis, M. A., K. N. K. Baker, D. R. Call, L. D. Warnick, Y. Soyer, M. Wiedmann, Y.
Grohn, P. L. McDonough, D. D. Hancock, and T. E. Besser. 2009. Multilocus variable-
number tandem-repeat method for typing Salmonella enterica serovar Newport. J. Clin.
Microbiol. 47:1934-1938.
14. De Reu, K., K. Grijspeerdt, W. Messens, M. Heyndrickx, M. Uyttendaele, J. Debevere,
and L. Herman. 2006. Eggshell factors influencing eggshell penetration and whole egg
contamination by different bacteria, including Salmonella Enteritidis. Int. J. Food Microbiol.
112:253-260.
111
15. Desai, M., E. J. Threlfall, and J. Stanley. 2001. Fluorescent amplified-fragment length
polymorphism subtyping of the Salmonella enterica serovar Enteritidis phage type 4 clone
complex. J. Clin. Microbiol. 39:201-206.
16. Deveau, H., R. Barrangou, J. E. Garneau, J. Labonte, C. Fremaux, P. Boyaval, D. A.
Romero, P. Horvath, and S. Moineau. 2007. Phage response to CRISPR-encoded resistance
in Streptococcus thermophilus. J. Bacteriol. 190:1390-1400.
17. Gast, R. K., and C. W. Beard. 1990. Production of Salmonella Enteritidis-contaminated
eggs by experimentally infected hens. Avian Dis. 34:438-446.
18. Gerner-Smidt, P., K. Hise, J. Kincaid, S. Hunter, S. Rolando, E. Hyytiä-Trees, E. M.
Ribot, B. Swaminathan, and Pulsenet Taskforce. 2006. Pulsenet USA: a five-year update.
Foodborne Pathog. Dis. 3:9-19.
19. Giammanco, G. M., C. Mammina, C. Romani, I. Luzzi, A. M. Dionisi, and A. Nastasi.
2007. Evaluation of a modified single-enzyme amplified fragment length polymorphism (SE-
AFLP) technique for subtyping Salmonella enterica serotype Enteritidis. Res. Microbiol.
158:10-17.
20. Grissa, I., G. Vergnaud, and C. Pourcel. 2008. CRISPRcompar: a website to compare
clustered regularly interspaced short palindromic repeats. Nucl. Acids Res. 36:W145-148.
21. Guard-Petter, J. 2001. The chicken, the egg and Salmonella Enteritidis. Environ. Microbiol.
3:421-430.
22. Kimura, A., V. Reddy, R. Marcus, P. Cieslak, J. Mohle‐Boetani, H. Kassenborg, S.
Segler, F. Hardnett, T. Barrett, and D. Swerdlow. 2004. Chicken consumption is a newly
identified risk factor for sporadic Salmonella enterica serotype Enteritidis infections in the
United States: a case‐control study in FoodNet sites. Clin. Infect. Dis. 38:S244-S252.
112
23. Kotetishvili, M., O. C. Stine, A. Kreger, J. G. Morris J., and A. Sulakvelidze. 2002.
Multilocus sequence typing for characterization of clinical and environmental Salmonella
strains. J. Clin. Microbiol. 40:1626-1635.
24. Kunin, V., S. He, F. Warnecke, B. S. Peterson, M. Haynes, N. Ivanova, L. L. Blackall, M.
Breitbart, F. Rohwer, D. Mcmahon, and P. Hugenholtz. 2008. A bacterial metapopulation
adapts locally to phage predation despite global dispersal. Genome Res. 18:293-297.
25. Lan, R., P. R. Reeves, and S. Octavia. 2009. Population structure, origins and evolution of
major Salmonella enterica clones. Infect. Genet. Evol. 9:996-1005.
26. Liebana, E., C. Clouting, L. Garcia-Migura, F. Clifton-Hadley, E. Lindsay, E. J.
Threlfall, and R. H. Davies. 2004. Multiple genetic typing of Salmonella Enteritidis phage-
types 4, 6, 7, 8 and 13a isolates from animals and humans in the UK. Vet. Microbiol.
100:189-195.
27. Lindstedt, B. -., E. Heir, T. Vardund, and G. Kapperud. 2000. A variation of the
amplified-fragment length polymorphism (AFLP) technique using three restriction
endonucleases, and assessment of the enzyme combination BglII-MfeI for AFLP analysis of
Salmonella enterica subsp. enterica isolates. FEMS Microbiol. Lett. 189:19-24.
28. Lindstedt, B., E. Heir, T. Vardund, and G. Kapperud. 2000. Fluorescent amplified-
fragment length polymorphism genotyping of Salmonella enterica subsp. enterica serovars
and comparison with pulsed-field gel electrophoresis typing. J. Clin. Microbiol. 38:1623-
1627.
29. Liu, F., R. Barrangou, P. Gerner-Smidt, E. Ribot, S. Knabel, and E. Dudley. Novel
virulence gene and CRISPR multilocus sequence typing scheme for subtyping the major
serovars of Salmonella enterica subspecies enterica . J. Clin. Microbiol., in press.
30. Marcus, R., J. K. Varma, C. Medus, E. J. Boothe, B. J. Anderson, T. Crume, K. E.
Fullerton, M. R. Moore, P. L. White, E. Lyszkowicz, A. C. Voetsch, and F. J. Angulo.
113
2006. Re-assessment of risk factors for sporadic Salmonella serotype Enteritidis infections: a
case-control study in five FoodNet sites, 2002–2003. Epidemiol. Infect. 7:1-9.
31. Okamura, M., Y. Kamijima, T. Miyamoto, H. Tani, K. Sasai, and E. Baba. 2001.
Differences among six Salmonella serovars in abilities to colonize reproductive organs and to
contaminate eggs in laying hens. Avian Dis. 45:61-69.
32. Pang, J., T. Chiu, R. Helmuth, A. Schroeter, B. Guerra, and H. Tsen. 2007. A pulsed
field gel electrophoresis (PFGE) study that suggests a major world-wide clone of Salmonella
enterica serovar Enteritidis. Int. J. Food Microbiol. 116:305-312.
33. Parker, C. T., S. Huynh, B. Quinones, L. J. Harris, and R. E. Mandrell. 2010.
Comparative genotypes of Salmonella enterica serovar Enteritidis phage type 30 and 9c
strains isolated from three outbreaks associated with raw almonds. Appl. Environ. Microbiol.
doi: 10.1128/AEM.03053-09.
34. Patrick, M. E., P. M. Adcock, T. M. Gomez, S. F. Altekruse, B. H. Holland, and R. V.
Tauxe. 2004. Salmonella Enteritidis Infections, United States, 1985–1999. Emerging Infect.
Dis. 10:1-7.
35. Ross, I. L., and M. W. Heuzenroeder. 2009. A comparison of two PCR-based typing
methods with pulsed-field gel electrophoresis in Salmonella enterica serovar Enteritidis. Int. J.
Med. Microbiol. 299:410-420.
36. Scott, F., J. Threlfall, J. Stanley, and C. Arnold. 2001. Fluorescent amplified fragment
length polymorphism genotyping of Salmonella Enteritidis: a method suitable for rapid
outbreak recognition. Clin. Microbiol. Infect. 7:479-485.
37. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: molecular evolutionary
genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24:1596-1599.
38. Torpdahl, M., M. N. Skov, D. Sandvang, and D. L. Baggesen. 2005. Genotypic
characterization of Salmonella by multilocus sequence typing, pulsed-field gel
114
electrophoresis and amplified fragment length polymorphism. J. Microbiol. Methods. 63:173-
184.
39. Touchon, M., and P. C. E. Rocha. 2010. The small, slow and specialized CRISPR and anti-
CRISPR of Escherichia and Salmonella. PLoS One. 5:e11126.
40. Vale, P. F., and T. J. Little. 2010. CRISPR-mediated phage resistance and the ghost of
coevolution past. Proc. R. Soc. Lond., B, Biol. Sci. 277: 2097-2103.
41. Voetsch, A., T. Van Gilder, F. Angulo, M. Farley, S. Shallow, R. Marcus, P. Cieslak, V.
Deneen, and R. Tauxe. 2004. FoodNet estimate of the burden of illness caused by
nontyphoidal Salmonella infections in the United States. Clin. Infect. Dis. 38:127-134.
42. Zheng, J., C. E. Keys, S. Zhao, J. Meng, and E. W. Brown. 2007. Enhanced subtyping
scheme for Salmonella Enteritidis. Emerging Infect. Dis.13:1932-1935.
115
Chapter 5
Conclusions and future research
5.1 Conclusions
In the current study, virulence genes identified three epidemic clones and one outbreak
clones of S. Enteritidis among clinical, poultry and environmental isolates. Moreover, this study
suggested that virulence genes may be used to identify strains that differ in virulence capacity,
which is very useful in terms of epidemiology. However, virulence genes could not separate
individual outbreak clones within an epidemic clone.
The present study suggested that CRISPRs are good epidemiological markers and may be
implemented for outbreak investigations. Addition of CRISPR sequences dramatically improved
discriminatory power of this MLST method by accurately differentiating individual outbreak
clones. CRISPRs are evolving due to plasmids and/or phage present in the environment, so
CRISPRs may reflect the specific phage and plasmid pool in the environment and hence contain
ecologically and geographically meaningful information for bacteria. As a result, CRISPRs may
be useful for tracing an outbreak clone of Salmonella to the specific farm or food processing plant
which serves as the reservoir for the source strain of an outbreak.
This study also showed that MLST based on virulence genes and CRISPRs has the
potential to be integrated into public surveillance laboratories to complement PFGE for
Salmonella outbreak investigations. First, the current MLST scheme accurately differentiated
outbreak clones of the major nine serovars of Salmonella. Second, the MLST scheme described
in the present study has superior discriminatory power, compared to previously published MLST
schemes for subtyping the major serovars of Salmonella. Third, the current MLST scheme
116
effectively subtyped the two most common PFGE patterns of S. Enteritidis and thus could
enhance cluster detection and outbreak investigation capabilities of this highly clonal serovar.
Four, the MLST scheme was highly congruent with serotyping for the major nine serovars of
Salmonella. To summerize, the MLST scheme described in the current study maybe an excellent
subtyping method for tracking the farm-to-fork spread of the major serovars of Salmonella during
outbreaks.
Additionally, the present MLST scheme based virulence genes and CRISPRs may be an
excellent tool to study the epidemiology of S. Enteritidis. MLST identified four STs (E ST1, 3, 4
and 8) that might represent four predominate and persistent STs circulating among humans in the
U.S. and poultry and hen house environments in PA. It also identified an environmental specific
sequence type E ST10, which might represent an environment-adapted clone of S. Enteritidis.
Moreover, cluster analysis based on fimH and sseL identified three epidemic clones and one
outbreak clone of S. Enteritidis, as well as 11 singletons. Significant differences in virulence
gene sequences and spacer arrangements between singletons and the other STs suggested that
singletons might have different virulence capacity than other STs.
In conclusion, the MLST scheme based on virulence genes and CRISPRs maybe an
excellent tool for subtyping this important foodborne pathogen during outbreak investigations.
Additionally, the present MLST scheme may be a valuable molecular subtyping method for
tracking the spread of S. Enteritidis throughout the poultry and egg food systems by providing
information about the ecological origin of S. Enteritidis isolates.
117
5.2 Future research
Although this MLST scheme shows great promise, future research is needed to further
validate it for molecular epidemiologic purposes. First, future research is needed using a random
collection of isolates representing a larger number of outbreaks, or otherwise epidemiologically
related isolates, to accurately compare the epidemiologic concordance of the present MLST
scheme with PFGE. A nonbiased strain collection needs to be included in future studies because
the isolates selected for this study were biased towards those that had poor epidemiologic
concordance of PFGE. Second, isolates from other major serovars of Salmonella should be tested
with the present MLST scheme. For instance, srovars Braenderup, Oranienburg, Agona, Infantis
and Thompson are of epidemiologic importance for causing many cases of human salmonellosis.
Third, future studies are required to investigate the ability of the present MLST scheme to
differentiate outbreak clones of S. Muenchen. Poor epidemiologic concordance for S. Muenchen
was observed in the present study and research is needed to test the following hypotheses: 1)
PFGE lacked discriminatory power for subtyping S. Muenchen; 2) CRISPRs involve too fast for
S. Muenchen.
Future research is required to investigate why the congruence between the current MLST
scheme and serotyping occurred. I speculate that the correlation between serotyping and the
MLST scheme results from the possible correlations between markers targeted by each subtyping
method. Serotyping targets the O antigens (lipopolysaccharide) and H antigens (peritrichous
flagella), which might involve in recognition and attachment to host cells. fimH gene encode
fimbrial adhesion which allows Salmonella to recognize and adhere to different receptors on host
cells, while sseL helps Salmonella to survive and replicate inside the host. CRISPRs defend
against phage and plasmids in the specific environment. These markers are all related in the
118
interaction of Salmonella with the host environment, which might explain the congruence
between serotyping and the present MLST scheme. To test this hypothesis, future research is
required.
Future studies are required to compare the current MLST scheme with Multiple Loci
VNTR Analysis (MLVA) for subtyping the highly clonal serovar S. Enteritidis. A number of
MLVA studies showed better discriminatory power than PFGE for subtyping S. Enteritidis.
Furthermore, MLVA has been used successfully along with other subtyping methods for S.
Enteritidis outbreak investigations. Therefore, comparison of discriminatory power and
epidemiologic concordance between MLVA and the current MLST scheme using a collection of
well-characterized isolates of S. Enteritidis from a number of outbreaks is needed.
Previous studies and the current study showed that PFGE banding patterns were not
stable during outbreak investigations, which suggested that the genomic content of Salmonella
might change during outbreaks. Likewise, how stable are CRISPRs during outbreaks? The
current study shown that CRISPRs appeared to be stable among isolates from the same outbreak;
however, future studies are needed to further test the stability of CRISPRs by mimicking the
farm-to-fork transmission of Salmonella during outbreaks and comparing CRISPRs in isolates
from different transmission points.
Moreover, future research is needed to examine how CRISPR1 and CRISPR2 evolve in
Salmonella. The present study suggested that the rate of spacer intake and deletion in CRISPRs
is suitable for Salmonella outbreak investigations. However, there were no studies that measured
the rate of CRISPR spacer intake and deletion. Additionally, future studies should investigate the
role of CRISPR1 and CRISPR2 in defending the bacteria against phage, respectively. The
present study observed more variability in spacer arrangements in CRISPR2 than CRISPR1 in
most serovars, except serovar S. Muenchen, in which CRISPR1 appeared to be more active than
119
CRISPR2. Studies are needed to unveil the reasons for these observations in order to better
understand the roles and mechanisms of CRISPRs in different serovars of Salmonella.
Additional research is also needed to determine the source of S. Enteritidis contamination
and the routes of transmission throughout the egg and poultry food system. First, the present
study only included poultry and environmental isolates from Pennsylvania, so it would be
interesting to include poultry isolates from other states in the U.S. Second, analysis of S.
Enteritidis isolates from breeder flocks in the future would help determine whether or not they
serve as the original source of S. Enteritidis sequence types causing illness in both poultry and
humans. Third, isolates from other potential contamination sources in the egg and poultry food
system should be characterized by the present MLST scheme. By comparing sequence types of
isolates from different sources along the production system with clinical sequence type, the
original contamination source and routes of transmission could be identified. Lastly, the present
study identified two major epidemic clones (ECI and ECII) in the U.S. It would be interesting to
investigate whether they are also major epidemic clones in the world. Previous studies have
observed the existence of major clones of S. Enteritidis in human and poultry in different
countries by PFGE. Therefore, future studies are required to characterize S. Enteritidis isolates
from other countries using the present MLST scheme.
Future experiments are required to compare the virulence of singletons with clinical
isolates. Data in the present study suggested that six singletons (E ST15, 21, 22, 23, 24 and 25)
may be adapted to poultry and only pathogenic to chickens and the other five singletons (E ST11,
16, 17, 19 and 20) might be non pathogenic to both humans and chickens. Future experiments
that examine the virulence of these singletons can test the above hypotheses. We also speculate
that E ST10, the environment-adapted sequence type, might have decreased virulence as well, and
this hypothesis should be tested in the future as well.
120
Future studies are required to investigate sequence type distributions of major serovars in
other animals that serve as reservoirs for Salmonella, such as turkey, cattle and swine.
Characterization of isolates from these animals by the current MLST scheme can broaden our
understanding of epidemiology of Salmonella in animals as a whole. Many questions need to be
answered. What are the contamination sources and routs of transmission of Salmonella in farms?
Are there major epidemic clones that circulate among different animals and causing most cases of
salmonellosis? Would sequence type distribution vary among different animals and are there
animal-specific sequence types?
Lastly, future studies are required to investigate virulence genes and CRISPRs as MLST
markers for subtyping other important foodborne pathogens. In the present study, MLST based
on virulence genes and CRISPRs allowed accurate identifications of outbreak clones for the
major serovars of Salmonella, so similar MLST schemes could be developed for other foodborne
pathogens. CRISPRs have been identified within the genomes of many bacterial species,
including many foodborne pathogens. For example, Campylobacter, Vibrio and Yersinia are
common foodborne pathogens causing many human infections and harbors CRISPR loci at the
same time. Development of MLST schemes based on virulence genes and CRISPRs for these
pathogens could potentially aid outbreak investigations and prevent future diseases.
121
APPENDIX
Supplemental materials
Table S1. Primers used to amplify and sequence other virulence genes
Marker Primer type1
Primer sequence (5'-3')
hilA Forward TTAATCGTCCGGTCGTAGTG
Reverse TCTGCCAGCGCACAGTAAGG
fimH2 Forward CCTCTTTTATTTGCTCTGCT
Reverse GTTATAAGCGAGGTCGTCAG
pipB Forward GGGCCTCTGTTTGAATACTT
Reverse ACAAAAATCACCTTATATCTTTTT
sopE Forward CGTCGCCATAAAAATGAATA
Reverse TGCATAGTTATCTAAAAGGAGAA
sseF Forward CGCAATCAAGATGAGTTATG
Reverse CACTCTCCATATTGGTTTCC
sseJ Forward CACTATGCCATTGAGTGTTG
Reverse ACCGGCACTATGATATTGAG
siiA Forward ATCAGGAGACAACATGGAAG
Reverse ATACCGGGAAAAGATAAAGC
sifB Forward TCGAATACCACCTATTCCAG
Reverse CAGGGGATTGTAAATCCATA
stdA Forward CAGGTATTTCAGGGTGTAGG
Reverse GTATGATGTATGGCGCTTCT
fimA Forward TATTGCGAGTCTGATGTTTG
Reverse TGACGGGATTATTCGTATTT
bcfC Forward TGCTTAAAAATATGGGGGTA
Reverse AAGGAAGGCTGTCGAATAAT
phoQ Forward CGATCCACAGTAAAGGAATG
Reverse TTGATAAAACCACCTTTCGT
1 Primers were used for both amplification and sequencing.
122
ST CRISPR1 CRISPR2
1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29 1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
33
33
SE str. P125109 p p p p p p p p p p p p p p p p p
EST3 p p p p p p p p p p p p p p p
EST2 p p p p p p p p p p p p p p p
EST9 p p p p p p p p p p p p p p p p
EST7 p p p p p p p p p p p p p p p p
EST1 p p p p p p p p p p p p p p p p p
EST8 p p p p p p p p p p p p p p p p p
EST6 p p p p p p p p p p p p p p p p p
EST4 p p p p p p p p p p p p p p p p p p p
EST5 p p p p p p p p p p p p p p p p
SN str. sl254 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
NST1 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
NST4 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
NST3 p p p p p p p p p p p p p p p p p p p p p p p
NST2 p p p p p p p p p p p p p p p p p p p
NST6 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
NST5 p p p p p p p p p p p p p p p p p p p p p p p p
ST str, LT2 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
TST7 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
TST1 p p p p p p p p p p p p p p p
TST2 p p p p p p p p p p p p p p p p p
TST3 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
TST6 p p p p p p p p p p p p p p p p p p p p p p p p p p
TST5 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
TST8 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
TST4 p p p p p p p p p p p p p p p p p p p p p p p
123
SH str. sl476 p p p p p p p p p p p p p p p p p p p p p v p p p p p p p p p p p p p p p p p p p p p
HST5 p p p p p p p p p p p p p p p p p p p p p p
HST4 p p p p p p p p p p p p p p p p p p p
HST1 p p p p p p p p p p p p p p p p p p p p p p p p p p
HST3 p p p p p p p p p p p p p p p p p p p p p p p p p p
HST2 p p p p p p p p p p p p p p p p p p p p p
HST6 p p p p p p p p p p p p p p p p p p p p
SS str. sara23 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
SST2 p p p p p p p p p p p p p p p p p p p p p p p p p p p p
SST1 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
SST3 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
SST4 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
SST5 p p p p p p p p p p p p p p p p p p p p p p p p p
SST6 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
SST8 p p p p p p p p p p p p p p p p p p p p p
JST7 p p p p p p p p p p p p p p p p p
JST1 p p p p p p p p p p p p p p p p p
JST8 p p p p p p p p p p p p p p p p p
JST5 p p p p p p p p p p p p p p p p p p p p p
JST3 p p p p p p p p p p p p p p p p p p p
JST9 p p p p p p p p p p p p p p
JST10 p p p p p p p p p
JST4 p p p p p p p p p p p p p p p p
JST6 p p p p p p p p p p p p p p p p p
JST2 p p p p p p
IST1 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
IST4 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
IST2 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
IST3 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
124
McnST7 p p p p p p p p
McnST2 p p p p p p p p p
McnST8 p p p p p p p p p p
McnST5 p p p p p p p p
McnST9 p p p p p p p p p p p p p p p p p p
McnST11 p p p p p p p p p p p p p p p p p p p p
McnST1 p p p p p p p p p p p p p p p p p p p p p p
McnST4 p p p p p p p p p p p p p p p p p p
McnST10 p p p p p p p p p p p p p p p p p p p
McnST6 p p p p p p p p p p p p p p p p p
McnST3 p p p p p p p p p p p p p p p p
McnST13 p p p p p p p p p p p p p p p p p p
McnST14 p p p p p p p p
McnST12 p p p p p p p p p p p p p p p p p
MvoST1 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
MvoST4 p p p p p p p p p p p p p p p p p p p p p p
MvoST5 p p p p p p p p p p p p p p p p p p p p p p p p
MvoST2 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
MvoST6 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p
MvoST3 p p p p p p p p p p p p p p p p p p p p p p p p l p p p p p p p p p p p p p p p
Figure S1. Graphic representation of spacer arrangements in CRISPR1 and CRISPR2.
Repeats are not included; only spacers are listed. The direction of the spacer arrangements is from the 5’ to 3’. Each unique
spacer is represented by a unique combination of the background color and the color of a particular character. Spacers were
aligned and gaps represent the absence of a particular spacer. Singletons had very distinct arrays of spacers and thus could not be
aligned. SE str. P125109, SN str. sl254, ST str, LT2, SH str. sl476 and SS str. sara23 are strains of S. Enteritidis, S. Newport, S.
Typhimurium, S. Heidelberg and S. Saintpaul, respectively, that have whole genome sequences deposited in GenBank.
125
Table S2. Source, isolate information and MLST results for the 167 isolates analyzed in the
present study
Code Source1 Sample type2 State Hen house Year Bird type MLST allelic profile
MLST ST3
fimH sseL CRISPR1 CRISPR2
CDC 1 CDC Clinical MN NA 2006 NA 1 1 1 1 E ST1
CDC 2 CDC Clinical MN NA 2006 NA 1 1 1 1 E ST1
CDC 3 CDC Clinical CA NA 2001 NA 1 2 2 2 E ST2
CDC 4 CDC Clinical CA NA 2001 NA 1 2 2 2 E ST2
CDC 5 CDC Clinical CA NA 2001 NA 1 2 2 2 E ST2
CDC 6 CDC Clinical ME NA 2006 NA 1 3 1 3 E ST3
CDC 7 CDC Clinical ME NA 2006 NA 1 3 1 3 E ST3
CDC 8 CDC Clinical PA NA 2007 NA 1 3 1 4 E ST4
CDC 9 CDC Clinical PA NA 2007 NA 1 3 1 4 E ST4
CDC 10 CDC Clinical GA NA 2005 NA 1 3 1 5 E ST5
CDC 11 CDC Clinical GA NA 2005 NA 1 3 1 4 E ST4
CDC 12 CDC Clinical ME NA 2006 NA 2 3 1 1 E ST6
CDC 13 CDC Clinical ME NA 2006 NA 2 3 1 1 E ST6
CDC 14 CDC Clinical ME NA 2006 NA 1 2 1 6 E ST7
CDC 15 CDC Clinical PA NA 2007 NA 1 3 1 4 E ST4
CDC 16 CDC Clinical GA NA 2005 NA 1 3 1 1 E ST8
CDC 17 CDC Clinical OH NA 2005 NA 1 1 1 6 E ST9
CDC 18 CDC Clinical MN NA 2008 NA 1 1 1 1 E ST1
CDC 19 CDC Clinical GA NA 2005 NA 1 3 1 1 E ST8
CDC 20 CDC Clinical OH NA 2005 NA 1 1 1 6 E ST9
CDC 21 CDC Clinical NA NA 2001 NA 1 2 2 2 E ST2
CDC 22 CDC Clinical OR NA 2005 NA 1 3 1 1 E ST8
CDC 23 CDC Clinical MN NA 2006 NA 1 1 1 1 E ST1
CDC 24 CDC Clinical WV NA 2009 NA 1 3 1 3 E ST3
CDC 25 CDC Clinical NA NA 2001 NA 1 2 2 2 E ST2
CDC 26 CDC Clinical CO NA 2008 NA 1 3 1 3 E ST3
CDC 27 CDC Clinical OR NA 2005 NA 1 3 1 1 E ST8
CDC 28 CDC Clinical SC NA 2005 NA 1 3 1 1 E ST8
CDC 29 CDC Clinical ID NA 2005 NA 1 3 1 1 E ST8
CDC 30 CDC Clinical GA NA 2005 NA 1 3 1 1 E ST8
CDC 31 CDC Clinical CO NA 2008 NA 1 3 1 3 E ST3
CDC 32 CDC Clinical MI NA 2007 NA 1 1 1 6 E ST9
CDC 33 CDC Clinical MI NA 2007 NA 1 1 1 6 E ST9
CDC 34 CDC Clinical CT NA 2007 NA 1 3 1 4 E ST4
PEQAP 76 PEQAP Env PA A 1998-1999 NA 1 3 1 19 E ST10
PEQAP 77 PEQAP Env PA E 1998-1999 NA 22 10 14 20 E ST11
PEQAP 78 PEQAP Env PA A 1998-1999 NA 1 3 1 19 E ST10
PEQAP 79 PEQAP Env PA B 1998-1999 NA 1 3 1 19 E ST10
PEQAP 80 PEQAP Env PA B 1998-1999 NA 1 3 1 19 E ST10
PEQAP 81 PEQAP Env PA A 1998-1999 NA 1 3 1 22 E ST14
PEQAP 82 PEQAP Env PA B 1998-1999 NA 1 3 1 6 E ST12
PEQAP 83 PEQAP Env PA F 1998-1999 NA 1 3 1 1 E ST8
PEQAP 84 PEQAP Env PA B 1998-1999 NA 1 3 1 19 E ST10
PEQAP 85 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 86 PEQAP Env PA C 1998-1999 NA 1 3 1 6 E ST12
PEQAP 87 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 88 PEQAP Egg PA B 1998-1999 NA 1 3 1 6 E ST12
PEQAP 89 PEQAP Env PA A 1998-1999 NA 1 3 1 6 E ST12
PEQAP 90 PEQAP Env PA A 1998-1999 NA 1 3 1 19 E ST10
PEQAP 91 PEQAP Egg PA B 1998-1999 NA 1 3 1 6 E ST12
PEQAP 92 PEQAP Env PA D 1998-1999 NA 1 3 1 1 E ST8
PEQAP 93 PEQAP Env PA E 1998-1999 NA 1 3 1 1 E ST8
PEQAP 94 PEQAP Env PA A 1998-1999 NA 1 3 1 19 E ST10
PEQAP 95 PEQAP Env PA NA 1998-1999 NA 1 3 1 19 E ST10
PEQAP 96 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 97 PEQAP Env PA E 1998-1999 NA 1 3 1 1 E ST8
126
PEQAP 98 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 99 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 100 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 101 PEQAP Env PA B 1998-1999 NA 1 3 1 6 E ST12
PEQAP 102 PEQAP Egg PA G 1998-1999 NA 1 3 1 6 E ST12
PEQAP 103 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 104 PEQAP Env PA NA 1998-1999 NA 1 2 15 6 E ST13
PEQAP 105 PEQAP Env PA NA 1998-1999 NA 1 3 1 6 E ST12
PEQAP 106 PEQAP Env PA NA 1998-1999 NA 9 11 16 23 E ST16
PEQAP 107 PEQAP Env PA NA 1998-1999 NA 10 12 17 24 E ST17
PEQAP 108 PEQAP Env PA NA 1998-1999 NA 1 3 1 25 E ST18
PEQAP 109 PEQAP Env PA NA 1998-1999 NA 1 3 1 25 E ST18
PEQAP 110 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 111 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 112 PEQAP Env PA NA 1998-1999 NA 11 13 18 26 E ST19
PEQAP 113 PEQAP Env PA NA 1998-1999 NA 1 3 1 19 E ST10
PEQAP 114 PEQAP Env PA NA 1998-1999 NA 1 3 1 19 E ST10
PEQAP 115 PEQAP Env PA NA 1998-1999 NA 12 14 19 27 E ST20
PEQAP 116 PEQAP Env PA NA 1998-1999 NA 1 3 1 19 E ST10
PEQAP 117 PEQAP Env PA NA 1998-1999 NA 1 3 1 19 E ST10
PEQAP 118 PEQAP Env PA NA 1998-1999 NA 1 3 1 19 E ST10
PEQAP 119 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 120 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 121 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 122 PEQAP Env PA NA 1998-1999 NA 1 3 1 1 E ST8
PEQAP 123 PEQAP Env PA NA 1998-1999 NA 1 1 1 1 E ST1
PEQAP 124 PEQAP Env PA NA 1998-1999 NA 1 1 1 1 E ST1
ADL 1 ADL Egg PA H 2008 Broiler 1 3 1 1 E ST8
ADL 2 ADL Egg PA I 2008 Broiler 1 1 1 1 E ST1
ADL 3 ADL Egg PA I 2008 Broiler 13 15 8 28 E ST21
ADL 4 ADL Egg PA I 2008 Broiler 1 1 1 1 E ST1
ADL 5 ADL Egg PA I 2008 Broiler 1 1 1 1 E ST1
ADL 6 ADL Egg PA I 2008 Broiler 13 15 10 28 E ST22
ADL 7 ADL Egg PA I 2008 Broiler 14 16 22 29 E ST23
ADL 8 ADL Egg PA I 2008 Broiler 1 1 1 1 E ST1
ADL 9 ADL Fecal PA NA 2008 Broiler 1 3 1 4 E ST4
ADL 10 ADL Necropsy PA J 2009 Broiler 1 3 1 4 E ST4
ADL 11 ADL Necropsy PA I 2009 Broiler 1 3 1 4 E ST4
ADL 12 ADL Egg PA K 2009 Broiler 1 3 1 4 E ST4
ADL 13 ADL Env PA K 2009 Broiler 1 3 1 4 E ST4
ADL 14 ADL Necropsy PA I 2009 Broiler 1 3 1 4 E ST4
ADL 15 ADL Necropsy PA I 2009 Broiler 15 17 6 30 E ST24
ADL 16 ADL Necropsy PA J 2009 Broiler 16 18 24 31 E ST25
ADL 17 ADL Egg PA I 2009 Broiler 1 3 1 4 E ST4
ADL 18 ADL Egg PA L 2007 Broiler 1 1 1 1 E ST1
ADL 19 ADL Egg PA L 2007 Broiler 1 1 1 1 E ST1
ADL 20 ADL Egg PA L 2007 Broiler 1 1 1 1 E ST1
ADL 21 ADL Egg PA L 2007 Broiler 1 1 1 1 E ST1
ADL 22 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 23 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 24 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 25 ADL Egg PA L 2007 Broiler 1 3 1 1 E ST8
ADL 26 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 27 ADL Egg PA I 2007 Broiler 1 1 1 1 E ST1
ADL 28 ADL Egg PA I 2007 Broiler 1 1 1 1 E ST1
ADL 29 ADL Egg PA I 2007 Broiler 1 3 1 1 E ST8
ADL 30 ADL Egg PA I 2007 Broiler 1 1 1 1 E ST1
ADL 31 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 32 ADL Egg PA I 2007 Broiler 1 1 1 1 E ST1
ADL 33 ADL Necropsy PA I 2007 Broiler 1 1 1 1 E ST1
ADL 34 ADL Egg PA L 2007 Broiler 1 1 1 1 E ST1
ADL 35 ADL Egg PA L 2007 Broiler 1 1 1 1 E ST1
127
ADL 36 ADL Necropsy PA I 2007 Broiler 1 3 1 4 E ST4
ADL 37 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 38 ADL Egg PA I 2007 Broiler 1 3 1 4 E ST4
ADL 39 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 40 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 41 ADL Egg PA I 2007 Broiler 1 1 1 1 E ST1
ADL 42 ADL Egg PA I 2007 Broiler 1 1 1 1 E ST1
ADL 43 ADL Egg PA L 2007 Broiler 1 3 1 4 E ST4
ADL 44 ADL Egg PA I 2007 Broiler 1 1 1 1 E ST1
ADL 45 ADL Egg PA L 2008 Broiler 1 3 1 4 E ST4
ADL 46 ADL Egg PA I 2008 Broiler 1 1 1 1 E ST1
ADL 47 ADL Egg PA I 2008 Broiler 1 1 1 1 E ST1
PEQAP 2 PEQAP Env PA M 2007 Layer 1 3 1 1 E ST8
PEQAP 3 PEQAP Egg PA M 2008 Layer 1 3 1 3 E ST3
PEQAP 4 PEQAP Egg PA M 2009 Layer 1 3 1 3 E ST3
PEQAP 5 PEQAP Egg PA M 2009 Layer 1 3 1 3 E ST3
PEQAP 6 PEQAP Env PA N 2008 Layer 1 1 1 1 E ST1
PEQAP 7 PEQAP Egg PA N 2008 Layer 1 1 1 1 E ST1
PEQAP 8 PEQAP Egg PA O 2007 Layer 1 3 1 4 E ST4
PEQAP 10 PEQAP Egg PA P 2009 Layer 1 3 1 1 E ST8
PEQAP 11 PEQAP Env PA P 2010 Layer 1 3 1 1 E ST8
PEQAP 12 PEQAP Egg PA P 2010 Layer 1 3 1 1 E ST8
PEQAP 13 PEQAP Env PA Q 2009 Layer 1 3 1 1 E ST8
PEQAP 14 PEQAP Egg PA Q 2009 Layer 1 3 1 1 E ST8
PEQAP 15 PEQAP Env PA Q 2010 Layer 1 3 1 1 E ST8
PEQAP 16 PEQAP Env PA R 2009 Layer 1 3 1 4 E ST4
PEQAP 18 PEQAP Egg PA S 2008 Layer 1 3 1 1 E ST8
PEQAP 19 PEQAP Env PA T 2009 Layer 1 3 1 22 E ST14
PEQAP22 PEQAP Env PA U 2007 Layer 1 3 1 22 E ST14
PEQAP23 PEQAP Env PA U 2007 Layer 1 3 1 71 E ST27
PEQAP25 PEQAP Egg PA V 2007 Layer 1 3 1 4 E ST4
PEQAP26 PEQAP Env PA W 2010 Layer 1 3 1 3 E ST3
PEQAP27 PEQAP Egg PA W 2010 Layer 1 3 1 3 E ST3
PEQAP28 PEQAP Egg PA W 2010 Layer 1 3 1 3 E ST3
PEQAP29 PEQAP Egg PA W 2010 Layer 1 3 1 3 E ST3
PEQAP30 PEQAP Env PA X 2010 Layer 1 3 1 1 E ST8
PEQAP31 PEQAP Egg PA X 2010 Layer 1 3 1 4 E ST4
PEQAP32 PEQAP Env PA Y 2010 Layer 1 3 1 1 E ST8
PEQAP33 PEQAP Egg PA Y 2010 Layer 17 19 65 73 E ST15
PEQAP34 PEQAP Env PA Z 2010 Layer 1 3 1 1 E ST8
PEQAP35 PEQAP Egg PA Z 2010 Layer 1 3 1 1 E ST8
PEQAP36 PEQAP Egg PA Z 2010 Layer 1 3 1 1 E ST8
PEQAP38 PEQAP Egg PA AA 2010 Layer 1 3 1 1 E ST8
PEQAP39 PEQAP Env PA AB 2008 Organic layer 1 3 1 1 E ST8
PEQAP40 PEQAP Env PA AB 2008 Organic layer 1 1 1 1 E ST1
PEQAP41 PEQAP Egg PA AB 2008 Organic layer 1 1 1 71 E ST26
PEQAP42 PEQAP Egg PA AB 2008 Organic layer 1 1 1 1 E ST1
PEQAP43 PEQAP Env PA AC 2008 Organic layer 1 1 1 1 E ST1
PEQAP44 PEQAP Egg PA AC 2010 Organic layer 1 1 1 1 E ST1
1 Isolates were obtained from CDC (Centers for Disease Control and Prevention), PEQAP
(Pennsylvania Egg Quality Assurance Program) and ADL (Animal Diagnostic Lab) in
Pennsylvania State University.
2 Sample type includes clinical, egg, necropsy and environmental isolates. Env stands for
environment.
128
3 ST: sequence type. E: S. Enteritidis. For instance, E ST1 stands for sequence type 1 for
Enteritidis.